id
string
sources
list
title
string
abstract
string
authors
list
categories
list
fields_of_study
list
published_date
timestamp[s]
url
string
pdf_url
string
arxiv_id
string
doi
string
citation_count
int64
influential_citation_count
int64
has_code
bool
code_url
string
venue
string
quality_score
float64
17788e19a5a78b04f17432163734845be109d6162adfdf0b579fcd4d610c1a95
[ "arxiv", "semantic_scholar" ]
Fisher Mask Nodes for Language Model Merging
Fine-tuning pre-trained models provides significant advantages in downstream performance. The ubiquitous nature of pre-trained models such as BERT and its derivatives in natural language processing has also led to a proliferation of task-specific fine-tuned models. As these models typically only perform one task well, ...
[ "Thennal D K", "Ganesh Nathan", "Suchithra M S" ]
[ "cs.CL", "cs.AI", "cs.LG" ]
[ "Computer Science" ]
2024-03-14T00:00:00
https://arxiv.org/abs/2403.09891
https://arxiv.org/pdf/2403.09891v3
2403.09891
10.48550/arXiv.2403.09891
9
0
false
null
International Conference on Language Resources and Evaluation
0.25
e12395dc31065654c16f4d000b45fe690fcf31589b8facdc5e5df13f6b6ce5ea
[ "arxiv", "semantic_scholar" ]
SemEval-2024 Shared Task 6: SHROOM, a Shared-task on Hallucinations and Related Observable Overgeneration Mistakes
This paper presents the results of the SHROOM, a shared task focused on detecting hallucinations: outputs from natural language generation (NLG) systems that are fluent, yet inaccurate. Such cases of overgeneration put in jeopardy many NLG applications, where correctness is often mission-critical. The shared task was c...
[ "Timothee Mickus", "Elaine Zosa", "Raúl Vázquez", "Teemu Vahtola", "Jörg Tiedemann", "Vincent Segonne", "Alessandro Raganato", "Marianna Apidianaki" ]
[ "cs.CL" ]
[ "Computer Science" ]
2024-03-12T00:00:00
https://arxiv.org/abs/2403.07726
https://arxiv.org/pdf/2403.07726v3
2403.07726
10.48550/arXiv.2403.07726
42
6
false
null
International Workshop on Semantic Evaluation
0.4225
5716178a6ce3a64b49b758de7ebe44d8aa43a51bebbb3ab5aee6710e73c37e47
[ "arxiv", "semantic_scholar" ]
A Segmentation Foundation Model for Diverse-type Tumors
Large pre-trained models with their numerous model parameters and extensive training datasets have shown excellent performance in various tasks. Many publicly available medical image datasets do not have a sufficient amount of data so there are few large-scale models in medical imaging. We propose a large-scale Tumor S...
[ "Jianhao Xie", "Ziang Zhang", "Guibo Luo", "Yuesheng Zhu" ]
[ "eess.IV", "cs.CV" ]
[ "Engineering", "Computer Science" ]
2024-03-11T00:00:00
https://arxiv.org/abs/2403.06396
https://arxiv.org/pdf/2403.06396v1
2403.06396
10.48550/arXiv.2403.06396
0
0
false
null
arXiv.org
0
3c504b3ad9a012ce3011989387018499c5d0ab82bb18c6ab6d7cfcf3567bd7bd
[ "arxiv", "semantic_scholar" ]
Training-Free Pretrained Model Merging
Recently, model merging techniques have surfaced as a solution to combine multiple single-talent models into a single multi-talent model. However, previous endeavors in this field have either necessitated additional training or fine-tuning processes, or require that the models possess the same pre-trained initializatio...
[ "Zhengqi Xu", "Ke Yuan", "Huiqiong Wang", "Yong Wang", "Mingli Song", "Jie Song" ]
[ "cs.CV" ]
[ "Computer Science" ]
2024-03-04T00:00:00
https://arxiv.org/abs/2403.01753
https://arxiv.org/pdf/2403.01753v3
2403.01753
10.1109/CVPR52733.2024.00565
36
0
true
https://github.com/zju-vipa/training_free_model_merging
Computer Vision and Pattern Recognition
0.3921
97bbc26f25c0cfecc3ce52bebc8455cd8694b30312be2db9f7969c9522ce0271
[ "arxiv", "semantic_scholar" ]
Merging Text Transformer Models from Different Initializations
Recent work on permutation-based model merging has shown impressive low- or zero-barrier mode connectivity between models from completely different initializations. However, this line of work has not yet extended to the Transformer architecture, despite its dominant popularity in the language domain. Therefore, in this...
[ "Neha Verma", "Maha Elbayad" ]
[ "cs.CL", "cs.AI", "cs.LG" ]
[ "Computer Science" ]
2024-03-01T00:00:00
https://arxiv.org/abs/2403.00986
https://arxiv.org/pdf/2403.00986v3
2403.00986
10.48550/arXiv.2403.00986
15
1
false
null
null
0.301
72f48530cbcd9beeb969f2ab0c42661455ff0d298a65e8591be8d7aa8def508e
[ "arxiv", "semantic_scholar" ]
Language Models are Homer Simpson! Safety Re-Alignment of Fine-tuned Language Models through Task Arithmetic
Aligned language models face a significant limitation as their fine-tuning often results in compromised safety. To tackle this, we propose a simple method RESTA that performs LLM safety realignment. RESTA stands for REstoring Safety through Task Arithmetic. At its core, it involves a simple arithmetic addition of a saf...
[ "Rishabh Bhardwaj", "Do Duc Anh", "Soujanya Poria" ]
[ "cs.CL", "cs.AI" ]
[ "Computer Science" ]
2024-02-19T00:00:00
https://arxiv.org/abs/2402.11746
https://arxiv.org/pdf/2402.11746v1
2402.11746
10.48550/arXiv.2402.11746
106
14
true
https://github.com/declare-lab/resta
Annual Meeting of the Association for Computational Linguistics
0.588
a1e3c7ec9b0520560ab9bd60709056519850bb92b6d4a63f65984a68f51b4726
[ "arxiv", "semantic_scholar" ]
WKVQuant: Quantizing Weight and Key/Value Cache for Large Language Models Gains More
Large Language Models (LLMs) face significant deployment challenges due to their substantial memory requirements and the computational demands of auto-regressive text generation process. This paper addresses these challenges by focusing on the quantization of LLMs, a technique that reduces memory consumption by convert...
[ "Yuxuan Yue", "Zhihang Yuan", "Haojie Duanmu", "Sifan Zhou", "Jianlong Wu", "Liqiang Nie" ]
[ "cs.LG", "cs.AI", "cs.CL" ]
[ "Computer Science" ]
2024-02-19T00:00:00
https://arxiv.org/abs/2402.12065
https://arxiv.org/pdf/2402.12065v2
2402.12065
10.48550/arXiv.2402.12065
85
3
false
null
arXiv.org
0.4836
c2e884c616d99c0b45de7f42f49d7c491805537ba6d607e924412c2438ec1a06
[ "arxiv", "semantic_scholar" ]
Learning From Failure: Integrating Negative Examples when Fine-tuning Large Language Models as Agents
Large language models (LLMs) have achieved success in acting as agents, which interact with environments through tools such as search engines. However, LLMs are optimized for language generation instead of tool use during training or alignment, limiting their effectiveness as agents. To resolve this problem, previous w...
[ "Renxi Wang", "Haonan Li", "Xudong Han", "Yixuan Zhang", "Timothy Baldwin" ]
[ "cs.CL" ]
[ "Computer Science" ]
2024-02-18T00:00:00
https://arxiv.org/abs/2402.11651
https://arxiv.org/pdf/2402.11651v2
2402.11651
10.48550/arXiv.2402.11651
46
4
false
null
arXiv.org
0.418
6dfa928b137534c5e038f8544820da357cf14b81f471f03c71ceeed777ab191f
[ "arxiv", "semantic_scholar" ]
Pelican Soup Framework: A Theoretical Framework for Language Model Capabilities
In this work, we propose a simple theoretical framework, Pelican Soup, aiming to better understand how pretraining allows LLMs to (1) generalize to unseen instructions and (2) perform in-context learning, even when the verbalizers are irrelevant to the task. To this end, in our framework, we introduce the notion of "kn...
[ "Ting-Rui Chiang", "Dani Yogatama" ]
[ "cs.CL", "cs.AI" ]
[ "Computer Science" ]
2024-02-16T00:00:00
https://arxiv.org/abs/2402.10424
https://arxiv.org/pdf/2402.10424v2
2402.10424
10.18653/v1/2026.findings-eacl.23
3
0
false
null
Conference of the European Chapter of the Association for Computational Linguistics
0.1505
7c68eda12ca260f0e2fe8896f5fb5567c8b2f57a8ec118ca9a8d5a19f564b530
[ "arxiv", "semantic_scholar" ]
Representation Surgery for Multi-Task Model Merging
Multi-task learning (MTL) compresses the information from multiple tasks into a unified backbone to improve computational efficiency and generalization. Recent work directly merges multiple independently trained models to perform MTL instead of collecting their raw data for joint training, greatly expanding the applica...
[ "Enneng Yang", "Li Shen", "Zhenyi Wang", "Guibing Guo", "Xiaojun Chen", "Xingwei Wang", "Dacheng Tao" ]
[ "cs.LG", "cs.AI", "cs.CV" ]
[ "Computer Science" ]
2024-02-05T00:00:00
https://arxiv.org/abs/2402.02705
https://arxiv.org/pdf/2402.02705v2
2402.02705
10.48550/arXiv.2402.02705
104
11
false
null
International Conference on Machine Learning
0.5396
7638c58ec905a473a4ea4112c7d4ef10a0947a2817b4a515efc3b12aadefa56d
[ "arxiv", "semantic_scholar" ]
Merging Multi-Task Models via Weight-Ensembling Mixture of Experts
Merging various task-specific Transformer-based models trained on different tasks into a single unified model can execute all the tasks concurrently. Previous methods, exemplified by task arithmetic, have been proven to be both effective and scalable. Existing methods have primarily focused on seeking a static optimal ...
[ "Anke Tang", "Li Shen", "Yong Luo", "Nan Yin", "Lefei Zhang", "Dacheng Tao" ]
[ "cs.LG", "cs.CV" ]
[ "Computer Science" ]
2024-02-01T00:00:00
https://arxiv.org/abs/2402.00433
https://arxiv.org/pdf/2402.00433v2
2402.00433
10.48550/arXiv.2402.00433
99
13
true
https://github.com/tanganke/weight-ensembling_MoE
International Conference on Machine Learning
0.5731
e8bd04765f835107f832036521f2a71da6fdc098b00987eb1f729ad8b42039a1
[ "arxiv", "semantic_scholar" ]
RADIN: Souping on a Budget
Model Soups, extending Stochastic Weights Averaging (SWA), combine models fine-tuned with different hyperparameters. Yet, their adoption is hindered by computational challenges due to subset selection issues. In this paper, we propose to speed up model soups by approximating soups performance using averaged ensemble lo...
[ "Thibaut Menes", "Olivier Risser-Maroix" ]
[ "cs.LG", "cs.CV" ]
[ "Computer Science" ]
2024-01-31T00:00:00
https://arxiv.org/abs/2401.17790
https://arxiv.org/pdf/2401.17790v1
2401.17790
10.48550/arXiv.2401.17790
1
0
false
null
arXiv.org
0.0753
9cbc2ae410bc3e48828c92d7ca8ef5165650e95657c3000c5c4bcd4ffdc7ca50
[ "arxiv", "semantic_scholar" ]
Active Inference as a Model of Agency
Is there a canonical way to think of agency beyond reward maximisation? In this paper, we show that any type of behaviour complying with physically sound assumptions about how macroscopic biological agents interact with the world canonically integrates exploration and exploitation in the sense of minimising risk and am...
[ "Lancelot Da Costa", "Samuel Tenka", "Dominic Zhao", "Noor Sajid" ]
[ "cs.AI" ]
[ "Computer Science" ]
2024-01-23T00:00:00
https://arxiv.org/abs/2401.12917
https://arxiv.org/pdf/2401.12917v1
2401.12917
10.48550/arXiv.2401.12917
16
2
false
null
arXiv.org
0.3076
12843b8c954813f221897042e5d835db3ef3aa5d4ffd6b6611b5d29eda95802d
[ "arxiv", "semantic_scholar" ]
CLIP Model for Images to Textual Prompts Based on Top-k Neighbors
Text-to-image synthesis, a subfield of multimodal generation, has gained significant attention in recent years. We propose a cost-effective approach for image-to-prompt generation that leverages generative models to generate textual prompts without the need for large amounts of annotated data. We divide our method into...
[ "Xin Zhang", "Xin Zhang", "YeMing Cai", "Tianzhi Jia" ]
[ "cs.CV", "cs.AI" ]
[ "Computer Science" ]
2024-01-18T00:00:00
https://arxiv.org/abs/2401.09763
https://arxiv.org/pdf/2401.09763v1
2401.09763
10.1109/EIECS59936.2023.10435489
2
0
false
null
null
0.1193
18fcca95943e51d2702adcd30900841dc0e750f405eeefdf754550b624b5021e
[ "arxiv", "semantic_scholar" ]
Erasing Undesirable Influence in Diffusion Models
Diffusion models are highly effective at generating high-quality images but pose risks, such as the unintentional generation of NSFW (not safe for work) content. Although various techniques have been proposed to mitigate unwanted influences in diffusion models while preserving overall performance, achieving a balance b...
[ "Jing Wu", "Trung Le", "Munawar Hayat", "Mehrtash Harandi" ]
[ "cs.CV" ]
[ "Computer Science" ]
2024-01-11T00:00:00
https://arxiv.org/abs/2401.05779
https://arxiv.org/pdf/2401.05779v4
2401.05779
10.1109/CVPR52734.2025.02632
41
6
false
null
Computer Vision and Pattern Recognition
0.4225
24418c7947f9ccffd4ab0a6c799f77800098fb070f1e012311313c976d9396a4
[ "arxiv", "semantic_scholar" ]
Merging Vision Transformers from Different Tasks and Domains
This work targets to merge various Vision Transformers (ViTs) trained on different tasks (i.e., datasets with different object categories) or domains (i.e., datasets with the same categories but different environments) into one unified model, yielding still good performance on each task or domain. Previous model mergin...
[ "Peng Ye", "Chenyu Huang", "Mingzhu Shen", "Tao Chen", "Yongqi Huang", "Yuning Zhang", "Wanli Ouyang" ]
[ "cs.CV", "cs.AI" ]
[ "Computer Science" ]
2023-12-25T00:00:00
https://arxiv.org/abs/2312.16240
https://arxiv.org/pdf/2312.16240v1
2312.16240
10.48550/arXiv.2312.16240
20
1
false
null
arXiv.org
0.3306
90bb504f9c9dabcbee76afa6327ad677fa167d68c23e0a099c905bcc21c45b9a
[ "arxiv", "semantic_scholar" ]
Model Breadcrumbs: Scaling Multi-Task Model Merging with Sparse Masks
The rapid development of AI systems has been greatly influenced by the emergence of foundation models. A common approach for targeted problems involves fine-tuning these pre-trained foundation models for specific target tasks, resulting in a rapid spread of models fine-tuned across a diverse array of tasks. This work f...
[ "MohammadReza Davari", "Eugene Belilovsky" ]
[ "cs.LG" ]
[ "Computer Science" ]
2023-12-11T00:00:00
https://arxiv.org/abs/2312.06795
https://arxiv.org/pdf/2312.06795v2
2312.06795
10.48550/arXiv.2312.06795
130
11
true
null
European Conference on Computer Vision
0.5396
8fc405edf05abccb4ed4920eba8116a92f73e9bd758c5deb8a34d5ef250687e2
[ "arxiv", "semantic_scholar" ]
Merging by Matching Models in Task Parameter Subspaces
Model merging aims to cheaply combine individual task-specific models into a single multitask model. In this work, we view past merging methods as leveraging different notions of a ''task parameter subspace'' in which models are matched before being merged. We connect the task parameter subspace of a given model to its...
[ "Derek Tam", "Mohit Bansal", "Colin Raffel" ]
[ "cs.LG", "cs.CL" ]
[ "Computer Science" ]
2023-12-07T00:00:00
https://arxiv.org/abs/2312.04339
https://arxiv.org/pdf/2312.04339v2
2312.04339
null
30
3
true
https://github.com/r-three/mats
null
0.3728
6ac5bc2b17c92fb1f9cae5098e3b4430b28e582c22fe804209d3ae5815ac5d53
[ "arxiv", "semantic_scholar" ]
Advances in the equivariant minimal model program and their applications in complex and arithmetic dynamics
This note reports some advances in the Equivariant Minimal Model Program (EMMP) for non-isomorphic surjective endomorphisms and their applications in complex and arithmetic dynamics.
[ "Sheng Meng", "De-Qi Zhang" ]
[ "math.AG", "math.DS", "math.NT" ]
[ "Mathematics" ]
2023-11-27T00:00:00
https://arxiv.org/abs/2311.16369
https://arxiv.org/pdf/2311.16369v1
2311.16369
10.1007/978-3-032-04048-0_4
8
4
false
null
DeMarco, L., Jonsson, M. (eds) Algebraic, Complex, and Arithmetic Dynamics. Simons Symposia. Springer, Cham. yr 2026, pages 99-123
0.3495
df1e874c85efbc5727e0229ca9923523e0a29352361e4ce75edabba001f445bd
[ "arxiv", "semantic_scholar" ]
Model Theory of Ultrafinitism II: Deconstructing the Term Model (First Draft)
This paper presents a novel possible worlds semantics, designed to elucidate the underpinnings of ultrafinitism. By constructing a careful modification of the well-known Kripke models for inuitionistic logic, we seek to extend our comprehension of the ultra-finite mindset. As it turns out, the passage from standard con...
[ "Mirco A. Mannucci" ]
[ "math.LO", "cs.LO" ]
[ "Mathematics", "Computer Science" ]
2023-11-26T00:00:00
https://arxiv.org/abs/2311.17931
https://arxiv.org/pdf/2311.17931v1
2311.17931
10.48550/arXiv.2311.17931
0
0
false
null
arXiv.org
0
48dd35cd0752349bfb25bba9eb2cf109a8349769b764f9b7d68b2c611b76b1d7
[ "arxiv", "semantic_scholar" ]
Orca 2: Teaching Small Language Models How to Reason
Orca 1 learns from rich signals, such as explanation traces, allowing it to outperform conventional instruction-tuned models on benchmarks like BigBench Hard and AGIEval. In Orca 2, we continue exploring how improved training signals can enhance smaller LMs' reasoning abilities. Research on training small LMs has often...
[ "Arindam Mitra", "Luciano Del Corro", "Shweti Mahajan", "Andres Codas", "Clarisse Simoes", "Sahaj Agarwal", "Xuxi Chen", "Anastasia Razdaibiedina", "Erik Jones", "Kriti Aggarwal", "Hamid Palangi", "Guoqing Zheng", "Corby Rosset", "Hamed Khanpour", "Ahmed Awadallah" ]
[ "cs.AI" ]
[ "Computer Science" ]
2023-11-18T00:00:00
https://arxiv.org/abs/2311.11045
https://arxiv.org/pdf/2311.11045v2
2311.11045
10.48550/arXiv.2311.11045
203
17
false
null
arXiv.org
0.6276
c73feb5738386fecd98077c0ba0bd3eb71297f63cb73d62cc32c5e53ce00b84b
[ "arxiv", "semantic_scholar" ]
PPTC Benchmark: Evaluating Large Language Models for PowerPoint Task Completion
Recent evaluations of Large Language Models (LLMs) have centered around testing their zero-shot/few-shot capabilities for basic natural language tasks and their ability to translate instructions into tool APIs. However, the evaluation of LLMs utilizing complex tools to finish multi-turn, multi-modal instructions in a c...
[ "Yiduo Guo", "Zekai Zhang", "Yaobo Liang", "Dongyan Zhao", "Nan Duan" ]
[ "cs.CL" ]
[ "Computer Science" ]
2023-11-03T00:00:00
https://arxiv.org/abs/2311.01767
https://arxiv.org/pdf/2311.01767v2
2311.01767
10.48550/arXiv.2311.01767
30
4
true
https://github.com/gydpku/PPTC}
Annual Meeting of the Association for Computational Linguistics
0.3728
f40bb48f8ff9401d0a44c442ec42131e4a5ceccda687c8cd30c7e15db8481d37
[ "arxiv", "semantic_scholar" ]
Model Merging by Uncertainty-Based Gradient Matching
Models trained on different datasets can be merged by a weighted-averaging of their parameters, but why does it work and when can it fail? Here, we connect the inaccuracy of weighted-averaging to mismatches in the gradients and propose a new uncertainty-based scheme to improve the performance by reducing the mismatch. ...
[ "Nico Daheim", "Thomas Möllenhoff", "Edoardo Maria Ponti", "Iryna Gurevych", "Mohammad Emtiyaz Khan" ]
[ "cs.LG", "cs.AI", "cs.CL" ]
[ "Computer Science" ]
2023-10-19T00:00:00
https://arxiv.org/abs/2310.12808
https://arxiv.org/pdf/2310.12808v2
2310.12808
10.48550/arXiv.2310.12808
90
7
true
https://github.com/UKPLab/iclr2024-model-merging
International Conference on Learning Representations
0.4898
3e2a7ecf50bf40e41986eb2987bc7740bfc76f6ff42dfa4d9975106a6b584a49
[ "arxiv", "semantic_scholar" ]
AdaMerging: Adaptive Model Merging for Multi-Task Learning
Multi-task learning (MTL) aims to empower a model to tackle multiple tasks simultaneously. A recent development known as task arithmetic has revealed that several models, each fine-tuned for distinct tasks, can be directly merged into a single model to execute MTL without necessitating a retraining process using the in...
[ "Enneng Yang", "Zhenyi Wang", "Li Shen", "Shiwei Liu", "Guibing Guo", "Xingwei Wang", "Dacheng Tao" ]
[ "cs.LG", "cs.CV" ]
[ "Computer Science" ]
2023-10-04T00:00:00
https://arxiv.org/abs/2310.02575
https://arxiv.org/pdf/2310.02575v2
2310.02575
10.48550/arXiv.2310.02575
251
49
false
null
International Conference on Learning Representations
0.8495
f9f5af4b05e69d84a46f2720a830267b7626e1098965fa1e851a5d3b98fa04d9
[ "arxiv", "semantic_scholar" ]
Soft Merging: A Flexible and Robust Soft Model Merging Approach for Enhanced Neural Network Performance
Stochastic Gradient Descent (SGD), a widely used optimization algorithm in deep learning, is often limited to converging to local optima due to the non-convex nature of the problem. Leveraging these local optima to improve model performance remains a challenging task. Given the inherent complexity of neural networks, t...
[ "Hao Chen", "Yusen Wu", "Phuong Nguyen", "Chao Liu", "Yelena Yesha" ]
[ "cs.LG" ]
[ "Computer Science" ]
2023-09-21T00:00:00
https://arxiv.org/abs/2309.12259
https://arxiv.org/pdf/2309.12259v1
2309.12259
10.48550/arXiv.2309.12259
0
0
false
null
arXiv.org
0
1228aed58d0055d19b8c959ff70dd01706e62c39d913f12f1e0e3648a1b174a6
[ "arxiv", "semantic_scholar" ]
Jais and Jais-chat: Arabic-Centric Foundation and Instruction-Tuned Open Generative Large Language Models
We introduce Jais and Jais-chat, new state-of-the-art Arabic-centric foundation and instruction-tuned open generative large language models (LLMs). The models are based on the GPT-3 decoder-only architecture and are pretrained on a mixture of Arabic and English texts, including source code in various programming langua...
[ "Neha Sengupta", "Sunil Kumar Sahu", "Bokang Jia", "Satheesh Katipomu", "Haonan Li", "Fajri Koto", "William Marshall", "Gurpreet Gosal", "Cynthia Liu", "Zhiming Chen", "Osama Mohammed Afzal", "Samta Kamboj", "Onkar Pandit", "Rahul Pal", "Lalit Pradhan", "Zain Muhammad Mujahid", "Mass...
[ "cs.CL", "cs.AI", "cs.LG" ]
[ "Computer Science" ]
2023-08-30T00:00:00
https://arxiv.org/abs/2308.16149
https://arxiv.org/pdf/2308.16149v2
2308.16149
10.48550/arXiv.2308.16149
81
8
false
null
arXiv.org
0.4785
0c4d4c25e58d3df50baf1e9e97d2ef83020f666d639d233208e3cc873bc3875e
[ "arxiv", "semantic_scholar" ]
Do the Frankenstein, or how to achieve better out-of-distribution performance with manifold mixing model soup
The standard recipe applied in transfer learning is to finetune a pretrained model on the task-specific dataset with different hyperparameter settings and pick the model with the highest accuracy on the validation dataset. Unfortunately, this leads to models which do not perform well under distribution shifts, e.g. whe...
[ "Hannes Fassold" ]
[ "cs.LG" ]
[ "Computer Science" ]
2023-08-28T00:00:00
https://arxiv.org/abs/2309.08610
https://arxiv.org/pdf/2309.08610v1
2309.08610
10.48550/arXiv.2309.08610
2
0
false
null
arXiv.org
0.1193
359a9f43b73be1ae8b68d02d70a11b2c929954c81d4809c08be6a4d172098efe
[ "arxiv", "semantic_scholar" ]
Testing different Log Bases For Vector Model Weighting Technique
Information retrieval systems retrieves relevant documents based on a query submitted by the user. The documents are initially indexed and the words in the documents are assigned weights using a weighting technique called TFIDF which is the product of Term Frequency (TF) and Inverse Document Frequency (IDF). TF represe...
[ "Kamel Assaf" ]
[ "cs.IR", "cs.AI" ]
[ "Computer Science" ]
2023-07-12T00:00:00
https://arxiv.org/abs/2307.06213
https://arxiv.org/pdf/2307.06213v1
2307.06213
10.5121/ijnlc.2023.12301
0
0
false
null
International Journal on Natural Language Computing
0
449368457b7e4ac759e65dcfcfc211dea651c7098e34835201caf6a17dddfc58
[ "arxiv", "semantic_scholar" ]
TBGC: Task-level Backbone-Oriented Gradient Clip for Multi-Task Foundation Model Learning
The AllInOne training paradigm squeezes a wide range of tasks into a unified model in a multi-task learning manner. However, optimization in multi-task learning is more challenge than single-task learning, as the gradient norm from different tasks may vary greatly, making the backbone overly biased towards one specific...
[ "Zelun Zhang", "Xue Pan" ]
[ "cs.CV", "cs.AI" ]
[ "Computer Science" ]
2023-07-07T00:00:00
https://arxiv.org/abs/2307.03465
https://arxiv.org/pdf/2307.03465v1
2307.03465
10.48550/arXiv.2307.03465
0
0
false
null
arXiv.org
0
8c1b1df24db248ecde782eddfc6095117f94be642a0d0cf3da7df321d92613d2
[ "arxiv", "semantic_scholar" ]
A Critical Look at the Current Usage of Foundation Model for Dense Recognition Task
In recent years large model trained on huge amount of cross-modality data, which is usually be termed as foundation model, achieves conspicuous accomplishment in many fields, such as image recognition and generation. Though achieving great success in their original application case, it is still unclear whether those fo...
[ "Shiqi Yang", "Atsushi Hashimoto", "Yoshitaka Ushiku" ]
[ "cs.CV" ]
[ "Computer Science" ]
2023-07-06T00:00:00
https://arxiv.org/abs/2307.02862
https://arxiv.org/pdf/2307.02862v2
2307.02862
10.48550/arXiv.2307.02862
1
0
false
null
arXiv.org
0.0753
5b041de7260623b1492c0988a7d3868b43ecb49875106769cd84c61c6de33980
[ "arxiv", "semantic_scholar" ]
Sparse Model Soups: A Recipe for Improved Pruning via Model Averaging
Neural networks can be significantly compressed by pruning, yielding sparse models with reduced storage and computational demands while preserving predictive performance. Model soups (Wortsman et al., 2022) enhance generalization and out-of-distribution (OOD) performance by averaging the parameters of multiple models i...
[ "Max Zimmer", "Christoph Spiegel", "Sebastian Pokutta" ]
[ "cs.LG", "cs.AI" ]
[ "Computer Science" ]
2023-06-29T00:00:00
https://arxiv.org/abs/2306.16788
https://arxiv.org/pdf/2306.16788v3
2306.16788
10.48550/arXiv.2306.16788
22
1
false
null
International Conference on Learning Representations
0.3404
fad81fd31133cd7516f98da301cae63fcab4654ed40279c46d410dbcc32115d1
[ "arxiv", "semantic_scholar" ]
Low-Rank Prune-And-Factorize for Language Model Compression
The components underpinning PLMs -- large weight matrices -- were shown to bear considerable redundancy. Matrix factorization, a well-established technique from matrix theory, has been utilized to reduce the number of parameters in PLM. However, it fails to retain satisfactory performance under moderate to high compres...
[ "Siyu Ren", "Kenny Q. Zhu" ]
[ "cs.CL" ]
[ "Computer Science" ]
2023-06-25T00:00:00
https://arxiv.org/abs/2306.14152
https://arxiv.org/pdf/2306.14152v1
2306.14152
10.48550/arXiv.2306.14152
19
1
false
null
International Conference on Language Resources and Evaluation
0.3253
2e4c5da564e29bd0fec7fba2ee0db41b2dafbdb3b61df2f0ff07f0f4688fc024
[ "arxiv", "semantic_scholar" ]
Rewarded soups: towards Pareto-optimal alignment by interpolating weights fine-tuned on diverse rewards
Foundation models are first pre-trained on vast unsupervised datasets and then fine-tuned on labeled data. Reinforcement learning, notably from human feedback (RLHF), can further align the network with the intended usage. Yet the imperfections in the proxy reward may hinder the training and lead to suboptimal results; ...
[ "Alexandre Ramé", "Guillaume Couairon", "Mustafa Shukor", "Corentin Dancette", "Jean-Baptiste Gaya", "Laure Soulier", "Matthieu Cord" ]
[ "cs.LG", "cs.AI", "cs.CV" ]
[ "Computer Science" ]
2023-06-07T00:00:00
https://arxiv.org/abs/2306.04488
https://arxiv.org/pdf/2306.04488v2
2306.04488
10.48550/arXiv.2306.04488
252
35
false
null
Neural Information Processing Systems
0.7782
8d45ab4b36ccb812b56bc774b27276e415acde39ec1875f3e0aa013eda05b006
[ "arxiv", "semantic_scholar" ]
TIES-Merging: Resolving Interference When Merging Models
Transfer learning - i.e., further fine-tuning a pre-trained model on a downstream task - can confer significant advantages, including improved downstream performance, faster convergence, and better sample efficiency. These advantages have led to a proliferation of task-specific fine-tuned models, which typically can on...
[ "Prateek Yadav", "Derek Tam", "Leshem Choshen", "Colin Raffel", "Mohit Bansal" ]
[ "cs.LG", "cs.AI", "cs.CL", "cs.CV" ]
[ "Computer Science" ]
2023-06-02T00:00:00
https://arxiv.org/abs/2306.01708
https://arxiv.org/pdf/2306.01708v2
2306.01708
10.52202/075280-0310
763
207
true
https://github.com/prateeky2806/ties-merging
Neural Information Processing Systems
1
3997689b467b637c16b76e9dd3e22e3bd84348a74367fef05e99a2e5c0e70069
[ "arxiv", "semantic_scholar" ]
Latent Diffusion Model Based Foley Sound Generation System For DCASE Challenge 2023 Task 7
Foley sound presents the background sound for multimedia content and the generation of Foley sound involves computationally modelling sound effects with specialized techniques. In this work, we proposed a system for DCASE 2023 challenge task 7: Foley Sound Synthesis. The proposed system is based on AudioLDM, which is a...
[ "Yi Yuan", "Haohe Liu", "Xubo Liu", "Xiyuan Kang", "Mark D. Plumbley", "Wenwu Wang" ]
[ "cs.SD", "cs.MM", "eess.AS" ]
[ "Computer Science", "Engineering" ]
2023-05-25T00:00:00
https://arxiv.org/abs/2305.15905
https://arxiv.org/pdf/2305.15905v3
2305.15905
10.48550/arXiv.2305.15905
11
2
false
null
arXiv.org
0.2698
7420aa9ccdb2f04bfd32a53a745bc0cc773b11a92c9e4374e0fd95f3d8f5219e
[ "arxiv", "semantic_scholar" ]
Adapting Language Models to Compress Contexts
Transformer-based language models (LMs) are powerful and widely-applicable tools, but their usefulness is constrained by a finite context window and the expensive computational cost of processing long text documents. We propose to adapt pre-trained LMs into AutoCompressors. These language models are capable of compress...
[ "Alexis Chevalier", "Alexander Wettig", "Anirudh Ajith", "Danqi Chen" ]
[ "cs.CL" ]
[ "Computer Science" ]
2023-05-24T00:00:00
https://arxiv.org/abs/2305.14788
https://arxiv.org/pdf/2305.14788v2
2305.14788
10.48550/arXiv.2305.14788
340
30
false
null
Conference on Empirical Methods in Natural Language Processing
0.7457
8955556fa2222a75dbf62343a5a1dee7cfe69be86edbe6ebfc960feb9af8fe95
[ "arxiv", "semantic_scholar" ]
Task Arithmetic in the Tangent Space: Improved Editing of Pre-Trained Models
Task arithmetic has recently emerged as a cost-effective and scalable approach to edit pre-trained models directly in weight space: By adding the fine-tuned weights of different tasks, the model's performance can be improved on these tasks, while negating them leads to task forgetting. Yet, our understanding of the eff...
[ "Guillermo Ortiz-Jimenez", "Alessandro Favero", "Pascal Frossard" ]
[ "cs.LG", "cs.CV" ]
[ "Computer Science" ]
2023-05-22T00:00:00
https://arxiv.org/abs/2305.12827
https://arxiv.org/pdf/2305.12827v3
2305.12827
10.48550/arXiv.2305.12827
221
16
false
null
Neural Information Processing Systems
0.6152
4dbeedd6e5b4a89c200ca4afad9b337ed039a4ee6d92cc4f67d1850032837765
[ "arxiv", "semantic_scholar" ]
MetaMorphosis: Task-oriented Privacy Cognizant Feature Generation for Multi-task Learning
With the growth of computer vision applications, deep learning, and edge computing contribute to ensuring practical collaborative intelligence (CI) by distributing the workload among edge devices and the cloud. However, running separate single-task models on edge devices is inefficient regarding the required computatio...
[ "Md Adnan Arefeen", "Zhouyu Li", "Md Yusuf Sarwar Uddin", "Anupam Das" ]
[ "cs.CV", "cs.CR", "cs.DC" ]
[ "Computer Science" ]
2023-05-13T00:00:00
https://arxiv.org/abs/2305.07815
https://arxiv.org/pdf/2305.07815v1
2305.07815
10.1145/3576842.3582372
0
0
false
null
International Conference on Internet-of-Things Design and Implementation
0
b2c950f56cf7e45ba34a0e041e7ac60db327e0fabd6cdb051d141c7ed740c04e
[ "arxiv", "semantic_scholar" ]
HACK: Learning a Parametric Head and Neck Model for High-fidelity Animation
Significant advancements have been made in developing parametric models for digital humans, with various approaches concentrating on parts such as the human body, hand, or face. Nevertheless, connectors such as the neck have been overlooked in these models, with rich anatomical priors often unutilized. In this paper, w...
[ "Longwen Zhang", "Zijun Zhao", "Xinzhou Cong", "Qixuan Zhang", "Shuqi Gu", "Yuchong Gao", "Rui Zheng", "Wei Yang", "Lan Xu", "Jingyi Yu" ]
[ "cs.GR" ]
[ "Computer Science" ]
2023-05-08T00:00:00
https://arxiv.org/abs/2305.04469
https://arxiv.org/pdf/2305.04469v1
2305.04469
10.1145/3592093
17
0
true
https://github.com/ZoneLikeWonderland/HACK-Model
ACM Transactions on Graphics
0.3138
947af38b39c5d4f1064239c6549c7181cc287e1c3a68090e3fd5c01b095e0e60
[ "arxiv", "semantic_scholar" ]
ZipIt! Merging Models from Different Tasks without Training
Typical deep visual recognition models are capable of performing the one task they were trained on. In this paper, we tackle the extremely difficult problem of combining distinct models with different initializations, each solving a separate task, into one multi-task model without any additional training. Prior work in...
[ "George Stoica", "Daniel Bolya", "Jakob Bjorner", "Pratik Ramesh", "Taylor Hearn", "Judy Hoffman" ]
[ "cs.CV", "cs.LG" ]
[ "Computer Science" ]
2023-05-04T00:00:00
https://arxiv.org/abs/2305.03053
https://arxiv.org/pdf/2305.03053v3
2305.03053
10.48550/arXiv.2305.03053
195
19
false
null
International Conference on Learning Representations
0.6505
088e45e71740d3b664934a2f93796c41ba02689c97245e1f9e0541c2ef39e815
[ "arxiv", "semantic_scholar" ]
Predict NAS Multi-Task by Stacking Ensemble Models using GP-NAS
Accurately predicting the performance of architecture with small sample training is an important but not easy task. How to analysis and train dataset to overcome overfitting is the core problem we should deal with. Meanwhile if there is the mult-task problem, we should also think about if we can take advantage of their...
[ "Ke Zhang" ]
[ "cs.LG", "cs.CV", "stat.AP", "stat.CO" ]
[ "Computer Science", "Mathematics" ]
2023-05-02T00:00:00
https://arxiv.org/abs/2305.01667
https://arxiv.org/pdf/2305.01667v1
2305.01667
10.48550/arXiv.2305.01667
0
0
false
null
arXiv.org
0
a0a3232939aed41524a12267dac1b66f5754adc771eff0e4b0a4840c328471d6
[ "arxiv", "semantic_scholar" ]
An Empirical Study of Multimodal Model Merging
Model merging (e.g., via interpolation or task arithmetic) fuses multiple models trained on different tasks to generate a multi-task solution. The technique has been proven successful in previous studies, where the models are trained on similar tasks and with the same initialization. In this paper, we expand on this co...
[ "Yi-Lin Sung", "Linjie Li", "Kevin Lin", "Zhe Gan", "Mohit Bansal", "Lijuan Wang" ]
[ "cs.CV", "cs.AI", "cs.CL", "cs.LG" ]
[ "Computer Science" ]
2023-04-28T00:00:00
https://arxiv.org/abs/2304.14933
https://arxiv.org/pdf/2304.14933v2
2304.14933
10.48550/arXiv.2304.14933
57
1
true
https://github.com/ylsung/vl-merging
Conference on Empirical Methods in Natural Language Processing
0.4409
a8abd931ab3d779ef8a4fcba309820cfe86a4e1a9412aa7b71918bf7aa1a1156
[ "arxiv", "semantic_scholar" ]
HuaTuo: Tuning LLaMA Model with Chinese Medical Knowledge
Large Language Models (LLMs), such as the LLaMA model, have demonstrated their effectiveness in various general-domain natural language processing (NLP) tasks. Nevertheless, LLMs have not yet performed optimally in biomedical domain tasks due to the need for medical expertise in the responses. In response to this chall...
[ "Haochun Wang", "Chi Liu", "Nuwa Xi", "Zewen Qiang", "Sendong Zhao", "Bing Qin", "Ting Liu" ]
[ "cs.CL" ]
[ "Computer Science" ]
2023-04-14T00:00:00
https://arxiv.org/abs/2304.06975
https://arxiv.org/pdf/2304.06975v1
2304.06975
10.48550/arXiv.2304.06975
290
25
true
https://github.com/SCIR-HI/Huatuo-Llama-Med-Chinese
arXiv.org
0.7075
05043c4892f1a9f7493c0a4f01f2d58b449fa9b8727ac4c843f3043dbe31e1ea
[ "arxiv", "semantic_scholar" ]
Adjust factor with volatility model using MAXFLAT low-pass filter and construct portfolio in China A share market
In the field of quantitative finance, volatility models, such as ARCH, GARCH, FIGARCH, SV, EWMA, play the key role in risk and portfolio management. Meanwhile, factor investing is more and more famous since mid of 20 century. CAPM, Fama French three factor model, Fama French five-factor model, MSCI Barra factor model a...
[ "Ke Zhang" ]
[ "q-fin.RM", "q-fin.ST" ]
[ "Economics" ]
2023-03-29T00:00:00
https://arxiv.org/abs/2304.04676
https://arxiv.org/pdf/2304.04676v2
2304.04676
null
1
0
false
null
null
0.0753
36a257c08aa10dfefe4939d4cdd4a01b23eea6ad4581143a61ec24473191c7bd
[ "arxiv", "semantic_scholar" ]
Merging Decision Transformers: Weight Averaging for Forming Multi-Task Policies
Recent work has shown the promise of creating generalist, transformer-based, models for language, vision, and sequential decision-making problems. To create such models, we generally require centralized training objectives, data, and compute. It is of interest if we can more flexibly create generalist policies by mergi...
[ "Daniel Lawson", "Ahmed H. Qureshi" ]
[ "cs.LG", "cs.AI" ]
[ "Computer Science" ]
2023-03-14T00:00:00
https://arxiv.org/abs/2303.07551
https://arxiv.org/pdf/2303.07551v3
2303.07551
10.1109/ICRA57147.2024.10610919
15
2
true
https://github.com/daniellawson9999/merging-decision-transformers
IEEE International Conference on Robotics and Automation
0.301
0731e37079725049d04fb30f9c08fbdf15dc29a128da801b1852ecdab8654b8e
[ "arxiv", "semantic_scholar" ]
Toward Defining a Domain Complexity Measure Across Domains
Artificial Intelligence (AI) systems planned for deployment in real-world applications frequently are researched and developed in closed simulation environments where all variables are controlled and known to the simulator or labeled benchmark datasets are used. Transition from these simulators, testbeds, and benchmark...
[ "Katarina Doctor", "Christine Task", "Eric Kildebeck", "Mayank Kejriwal", "Lawrence Holder", "Russell Leong" ]
[ "cs.AI" ]
[ "Computer Science" ]
2023-03-07T00:00:00
https://arxiv.org/abs/2303.04141
https://arxiv.org/pdf/2303.04141v1
2303.04141
10.48550/arXiv.2303.04141
10
2
false
null
arXiv.org
0.2603
7642ac038b3bc81b3b6c1b1ea1d0d0856133425cc041d6bdcd7b62946d2cc230
[ "arxiv", "semantic_scholar" ]
Hitachi at SemEval-2023 Task 3: Exploring Cross-lingual Multi-task Strategies for Genre and Framing Detection in Online News
This paper explains the participation of team Hitachi to SemEval-2023 Task 3 "Detecting the genre, the framing, and the persuasion techniques in online news in a multi-lingual setup.'' Based on the multilingual, multi-task nature of the task and the low-resource setting, we investigated different cross-lingual and mult...
[ "Yuta Koreeda", "Ken-ichi Yokote", "Hiroaki Ozaki", "Atsuki Yamaguchi", "Masaya Tsunokake", "Yasuhiro Sogawa" ]
[ "cs.CL", "cs.AI" ]
[ "Computer Science" ]
2023-03-03T00:00:00
https://arxiv.org/abs/2303.01794
https://arxiv.org/pdf/2303.01794v2
2303.01794
10.18653/v1/2023.semeval-1.237
3
0
false
null
International Workshop on Semantic Evaluation
0.1505
8726ddc7761ff438489e794d0f1ddfa171f7211f5fc0e1a28712e0ae622a8acf
[ "arxiv", "semantic_scholar" ]
Seasoning Model Soups for Robustness to Adversarial and Natural Distribution Shifts
Adversarial training is widely used to make classifiers robust to a specific threat or adversary, such as $\ell_p$-norm bounded perturbations of a given $p$-norm. However, existing methods for training classifiers robust to multiple threats require knowledge of all attacks during training and remain vulnerable to unsee...
[ "Francesco Croce", "Sylvestre-Alvise Rebuffi", "Evan Shelhamer", "Sven Gowal" ]
[ "cs.LG", "cs.CV" ]
[ "Computer Science" ]
2023-02-20T00:00:00
https://arxiv.org/abs/2302.10164
https://arxiv.org/pdf/2302.10164v1
2302.10164
10.1109/CVPR52729.2023.01185
23
2
false
null
Computer Vision and Pattern Recognition
0.3451
60a8eb09320628f095ae2ce7a720d9ef1231ca8665bb3576e769ab6bb1c6261f
[ "arxiv", "semantic_scholar" ]
Zilber's notion of logically perfect structure: Universal Covers
We sketch recent interactions between model theory and a roughly 150-year old study of analytic functions involving complex analysis, algebraic topology, and number theory, centered in canonicity of universal covers. Towards this goal we discuss in a systematic and unified way several examples indicating the main ideas...
[ "John T. Baldwin", "Andrés Villaveces" ]
[ "math.LO" ]
[ "Mathematics" ]
2023-02-09T00:00:00
https://arxiv.org/abs/2302.04650
https://arxiv.org/pdf/2302.04650v3
2302.04650
10.2140/mt.2024.3.647
7
1
false
null
Model Th. 3 (2024) 647-683
0.2258
16776bc8de2ca19bcb0882e78081e34a989d813fc5e8325d2e6ec73b1b976eda
[ "arxiv", "semantic_scholar" ]
Backward Compatibility During Data Updates by Weight Interpolation
Backward compatibility of model predictions is a desired property when updating a machine learning driven application. It allows to seamlessly improve the underlying model without introducing regression bugs. In classification tasks these bugs occur in the form of negative flips. This means an instance that was correct...
[ "Raphael Schumann", "Elman Mansimov", "Yi-An Lai", "Nikolaos Pappas", "Xibin Gao", "Yi Zhang" ]
[ "cs.LG", "cs.CL" ]
[ "Computer Science" ]
2023-01-25T00:00:00
https://arxiv.org/abs/2301.10546
https://arxiv.org/pdf/2301.10546v1
2301.10546
10.48550/arXiv.2301.10546
7
1
false
null
Conference of the European Chapter of the Association for Computational Linguistics
0.2258
5612f4b4ebf284d6b2ac1bbcde9390a4ecf018a47ca2bb4b11dc879d552f0dd2
[ "arxiv", "semantic_scholar" ]
Dataless Knowledge Fusion by Merging Weights of Language Models
Fine-tuning pre-trained language models has become the prevalent paradigm for building downstream NLP models. Oftentimes fine-tuned models are readily available but their training data is not, due to data privacy or intellectual property concerns. This creates a barrier to fusing knowledge across individual models to y...
[ "Xisen Jin", "Xiang Ren", "Daniel Preotiuc-Pietro", "Pengxiang Cheng" ]
[ "cs.CL", "cs.LG" ]
[ "Computer Science" ]
2022-12-19T00:00:00
https://arxiv.org/abs/2212.09849
https://arxiv.org/pdf/2212.09849v6
2212.09849
10.48550/arXiv.2212.09849
388
69
true
https://github.com/bloomberg/dataless-model-merging
International Conference on Learning Representations
0.9225
385817bdb1c35abf723aac3315f39e9ebe64c685ee7fb704299e9f4693357e6b
[ "arxiv", "semantic_scholar" ]
Editing Models with Task Arithmetic
Changing how pre-trained models behave -- e.g., improving their performance on a downstream task or mitigating biases learned during pre-training -- is a common practice when developing machine learning systems. In this work, we propose a new paradigm for steering the behavior of neural networks, centered around \texti...
[ "Gabriel Ilharco", "Marco Tulio Ribeiro", "Mitchell Wortsman", "Suchin Gururangan", "Ludwig Schmidt", "Hannaneh Hajishirzi", "Ali Farhadi" ]
[ "cs.LG", "cs.CL", "cs.CV" ]
[ "Computer Science" ]
2022-12-08T00:00:00
https://arxiv.org/abs/2212.04089
https://arxiv.org/pdf/2212.04089v3
2212.04089
10.48550/arXiv.2212.04089
1,041
279
false
null
International Conference on Learning Representations
1
b00efef6a6fc88cfc94f229aca06d4c66cf649f4e3e3fa544d8a51d2a1e85044
[ "arxiv", "semantic_scholar" ]
Task-Driven Hybrid Model Reduction for Dexterous Manipulation
In contact-rich tasks, like dexterous manipulation, the hybrid nature of making and breaking contact creates challenges for model representation and control. For example, choosing and sequencing contact locations for in-hand manipulation, where there are thousands of potential hybrid modes, is not generally tractable. ...
[ "Wanxin Jin", "Michael Posa" ]
[ "cs.RO", "eess.SY" ]
[ "Computer Science", "Engineering" ]
2022-11-30T00:00:00
https://arxiv.org/abs/2211.16657
https://arxiv.org/pdf/2211.16657v2
2211.16657
10.1109/TRO.2024.3359531
21
2
true
https://github.com/wanxinjin/Task-Driven-Hybrid-Reduction
IEEE Transactions on robotics
0.3356
664a87f57af3cd5d180d4ee81431f7a09d03fae9adf2b4f1c1c7b86e66ce3aef
[ "arxiv", "semantic_scholar" ]
Modelling COVID-19-III: endemic spread in India
A disease in a given population is termed endemic when it exhibits a steady prevalence. We address the pertinent question as to what extent COVID-19 has turned endemic in India. There are several existing models for studying endemic behaviour, such as the extensions of the traditional temporal SIR model or the spatio-t...
[ "Madhuchhanda Bhattacharjee", "Arup Bose" ]
[ "stat.AP" ]
[ "Mathematics" ]
2022-11-11T00:00:00
https://arxiv.org/abs/2211.06215
https://arxiv.org/pdf/2211.06215v1
2211.06215
null
0
0
false
null
null
0
8248004dd3f126e04246c8c4e128c2453f2940ee29c04b0224804fb896241438
[ "arxiv", "semantic_scholar" ]
Momentum-based Weight Interpolation of Strong Zero-Shot Models for Continual Learning
Large pre-trained, zero-shot capable models have shown considerable success both for standard transfer and adaptation tasks, with particular robustness towards distribution shifts. In addition, subsequent fine-tuning can considerably improve performance on a selected downstream task. However, through naive fine-tuning,...
[ "Zafir Stojanovski", "Karsten Roth", "Zeynep Akata" ]
[ "cs.LG", "cs.CV" ]
[ "Computer Science" ]
2022-11-06T00:00:00
https://arxiv.org/abs/2211.03186
https://arxiv.org/pdf/2211.03186v1
2211.03186
10.48550/arXiv.2211.03186
19
3
false
null
arXiv.org
0.3253
2e7aefcc61141b7d9ec6d2b493c3fb26f50645a1a99342c1abac51eff834fa28
[ "arxiv", "semantic_scholar" ]
Where to start? Analyzing the potential value of intermediate models
Previous studies observed that finetuned models may be better base models than the vanilla pretrained model. Such a model, finetuned on some source dataset, may provide a better starting point for a new finetuning process on a desired target dataset. Here, we perform a systematic analysis of this intertraining scheme, ...
[ "Leshem Choshen", "Elad Venezian", "Shachar Don-Yehia", "Noam Slonim", "Yoav Katz" ]
[ "cs.CL", "cs.AI", "cs.LG" ]
[ "Computer Science" ]
2022-10-31T00:00:00
https://arxiv.org/abs/2211.00107
https://arxiv.org/pdf/2211.00107v3
2211.00107
10.48550/arXiv.2211.00107
30
2
false
null
Conference on Empirical Methods in Natural Language Processing
0.3728
dc86c27cffd75f86396f7e7c59ab719309678c0a3cf9c4c049fe04002b0ae906
[ "arxiv", "semantic_scholar" ]
Higher internal covers
We define and study a higher-dimensional version of model theoretic internality, and relate it to higher-dimensional definable groupoids in the base theory.
[ "Moshe Kamensky" ]
[ "math.LO", "math.CT" ]
[ "Mathematics" ]
2022-10-06T00:00:00
https://arxiv.org/abs/2210.02699
https://arxiv.org/pdf/2210.02699v2
2210.02699
10.2140/mt.2023.2.449
0
0
false
null
Model Th. 2 (2023) 449-479
0
b0b121dda62409ac21c90c941532e2c08311f9e0f0f701d2140e50254f2ec97f
[ "arxiv", "semantic_scholar" ]
Fractional Gagliardo-Nirenberg interpolation inequality and bounded mean oscillation
We prove Gagliardo-Nirenberg interpolation inequalities estimating the Sobolev semi-norm in terms of the bounded mean oscillation semi-norm and a Sobolev semi-norm, with some of the Sobolev semi-norms having fractional order.
[ "Jean Van Schaftingen" ]
[ "math.CA" ]
[ "Mathematics" ]
2022-08-31T00:00:00
https://arxiv.org/abs/2208.14691
https://arxiv.org/pdf/2208.14691v3
2208.14691
10.5802/crmath.463
7
1
false
null
Comptes rendus. Mathematique
0.2258
209a14ff922a2c84ba3ac9661679d1c0312bedc75f0f6c286cc050a8d4ac810b
[ "arxiv", "semantic_scholar" ]
Patching open-vocabulary models by interpolating weights
Open-vocabulary models like CLIP achieve high accuracy across many image classification tasks. However, there are still settings where their zero-shot performance is far from optimal. We study model patching, where the goal is to improve accuracy on specific tasks without degrading accuracy on tasks where performance i...
[ "Gabriel Ilharco", "Mitchell Wortsman", "Samir Yitzhak Gadre", "Shuran Song", "Hannaneh Hajishirzi", "Simon Kornblith", "Ali Farhadi", "Ludwig Schmidt" ]
[ "cs.CV", "cs.LG" ]
[ "Computer Science" ]
2022-08-10T00:00:00
https://arxiv.org/abs/2208.05592
https://arxiv.org/pdf/2208.05592v2
2208.05592
10.48550/arXiv.2208.05592
215
38
false
null
Neural Information Processing Systems
0.7955
ff130d892375d56be05f79909fc7e00952cfd2a5c0922e2b053f35c4bf061a8b
[ "arxiv", "semantic_scholar" ]
Understanding Weight Similarity of Neural Networks via Chain Normalization Rule and Hypothesis-Training-Testing
We present a weight similarity measure method that can quantify the weight similarity of non-convex neural networks. To understand the weight similarity of different trained models, we propose to extract the feature representation from the weights of neural networks. We first normalize the weights of neural networks by...
[ "Guangcong Wang", "Guangrun Wang", "Wenqi Liang", "Jianhuang Lai" ]
[ "cs.LG", "cs.CV", "math.ST", "stat.ML" ]
[ "Computer Science", "Mathematics" ]
2022-08-08T00:00:00
https://arxiv.org/abs/2208.04369
https://arxiv.org/pdf/2208.04369v1
2208.04369
10.48550/arXiv.2208.04369
5
0
false
null
arXiv.org
0.1945
8a79b13f689f78e400c93031b3d7e6719043e21a2972269548090db20c3b28c6
[ "arxiv", "semantic_scholar" ]
On the Usability of Transformers-based models for a French Question-Answering task
For many tasks, state-of-the-art results have been achieved with Transformer-based architectures, resulting in a paradigmatic shift in practices from the use of task-specific architectures to the fine-tuning of pre-trained language models. The ongoing trend consists in training models with an ever-increasing amount of ...
[ "Oralie Cattan", "Christophe Servan", "Sophie Rosset" ]
[ "cs.CL", "cs.AI" ]
[ "Computer Science" ]
2022-07-19T00:00:00
https://arxiv.org/abs/2207.09150
https://arxiv.org/pdf/2207.09150v1
2207.09150
10.26615/978-954-452-072-4_029
15
1
false
null
Recent Advances in Natural Language Processing
0.301
fa6a6521d44fe13d6307d030e24a9a86f209823f2a589f2ed59826e6ab4229af
[ "arxiv", "semantic_scholar" ]
A SIQRB delayed model for cholera and optimal control treatment
We improve a recent mathematical model for cholera by adding a time delay that represents the time between the instant at which an individual becomes infected and the instant at which he begins to have symptoms of cholera disease. We prove that the delayed cholera model is biologically meaningful and analyze the local ...
[ "Ana P. Lemos-Paiao", "Helmut Maurer", "Cristiana J. Silva", "Delfim F. M. Torres" ]
[ "math.OC", "q-bio.PE" ]
[ "Mathematics", "Biology" ]
2022-06-25T00:00:00
https://arxiv.org/abs/2206.12688
https://arxiv.org/pdf/2206.12688v1
2206.12688
10.1051/mmnp/2022027
14
1
false
null
Mathematical Modelling of Natural Phenomena
0.294
fb3be48833eea97af148d4982b502c8b89c5f1d155f2254c30816a1e37c07a7e
[ "arxiv", "semantic_scholar" ]
Classical Aspects of a Distributional 3+1 Foam Model
A 3+1 spacetime, with a shift vector that is the unique fundamental solution to the linearized wave operator, is introduced to model an interpretation of Wheeler's layman's analogy of the Quantum foam. To understand the distributional aspects of this model is the guaranteed existence of a sequence of compactly supporte...
[ "Claes Cramer" ]
[ "gr-qc", "math-ph", "quant-ph" ]
[ "Physics", "Mathematics" ]
2022-06-21T00:00:00
https://arxiv.org/abs/2206.10417
https://arxiv.org/pdf/2206.10417v11
2206.10417
null
0
0
false
null
null
0
e8dc77fa2f412f46222a98d4cad80854faf2f6c7cc6941e94f3e8cf7537af00c
[ "arxiv", "semantic_scholar" ]
UMass PCL at SemEval-2022 Task 4: Pre-trained Language Model Ensembles for Detecting Patronizing and Condescending Language
Patronizing and condescending language (PCL) is everywhere, but rarely is the focus on its use by media towards vulnerable communities. Accurately detecting PCL of this form is a difficult task due to limited labeled data and how subtle it can be. In this paper, we describe our system for detecting such language which ...
[ "David Koleczek", "Alex Scarlatos", "Siddha Karakare", "Preshma Linet Pereira" ]
[ "cs.CL" ]
[ "Computer Science" ]
2022-04-18T00:00:00
https://arxiv.org/abs/2204.08304
https://arxiv.org/pdf/2204.08304v1
2204.08304
10.48550/arXiv.2204.08304
1
0
false
null
International Workshop on Semantic Evaluation
0.0753
c17294dfc20d954ae895fc434ed6cdff85067bdba21dd67ff80f9351543a79ae
[ "arxiv", "semantic_scholar" ]
HFL at SemEval-2022 Task 8: A Linguistics-inspired Regression Model with Data Augmentation for Multilingual News Similarity
This paper describes our system designed for SemEval-2022 Task 8: Multilingual News Article Similarity. We proposed a linguistics-inspired model trained with a few task-specific strategies. The main techniques of our system are: 1) data augmentation, 2) multi-label loss, 3) adapted R-Drop, 4) samples reconstruction wit...
[ "Zihang Xu", "Ziqing Yang", "Yiming Cui", "Zhigang Chen" ]
[ "cs.CL" ]
[ "Computer Science" ]
2022-04-11T00:00:00
https://arxiv.org/abs/2204.04844
https://arxiv.org/pdf/2204.04844v1
2204.04844
10.48550/arXiv.2204.04844
8
2
false
null
International Workshop on Semantic Evaluation
0.2386
723e6d1888f5f93fdaf9d4bf96f8e528734ea47fb4d5755c357480de95f6abc7
[ "arxiv", "semantic_scholar" ]
Long-Tailed Recognition via Weight Balancing
In the real open world, data tends to follow long-tailed class distributions, motivating the well-studied long-tailed recognition (LTR) problem. Naive training produces models that are biased toward common classes in terms of higher accuracy. The key to addressing LTR is to balance various aspects including data distri...
[ "Shaden Alshammari", "Yu-Xiong Wang", "Deva Ramanan", "Shu Kong" ]
[ "cs.CV" ]
[ "Computer Science" ]
2022-03-27T00:00:00
https://arxiv.org/abs/2203.14197
https://arxiv.org/pdf/2203.14197v1
2203.14197
10.1109/CVPR52688.2022.00677
196
28
true
https://github.com/ShadeAlsha/LTR-weight-balancing
Computer Vision and Pattern Recognition
0.7312
ee63389abbb5e6afd52ce1bcc42f2ccb1eeac58164774d6395e3c69feced84ce
[ "arxiv", "semantic_scholar" ]
Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time
The conventional recipe for maximizing model accuracy is to (1) train multiple models with various hyperparameters and (2) pick the individual model which performs best on a held-out validation set, discarding the remainder. In this paper, we revisit the second step of this procedure in the context of fine-tuning large...
[ "Mitchell Wortsman", "Gabriel Ilharco", "Samir Yitzhak Gadre", "Rebecca Roelofs", "Raphael Gontijo-Lopes", "Ari S. Morcos", "Hongseok Namkoong", "Ali Farhadi", "Yair Carmon", "Simon Kornblith", "Ludwig Schmidt" ]
[ "cs.LG", "cs.CL", "cs.CV" ]
[ "Computer Science" ]
2022-03-10T00:00:00
https://arxiv.org/abs/2203.05482
https://arxiv.org/pdf/2203.05482v3
2203.05482
10.48550/arXiv.2203.05482
1,565
192
true
https://github.com/mlfoundations/model-soups
International Conference on Machine Learning
1
e8f60c07e4b269ece44a7009a2e7e90aaffd5c4668b2c6b687324b511ef284fe
[ "arxiv", "semantic_scholar" ]
Triple Motion Estimation and Frame Interpolation based on Adaptive Threshold for Frame Rate Up-Conversion
In this paper, we propose a novel motion-compensated frame rate up-conversion (MC-FRUC) algorithm. The proposed algorithm creates interpolated frames by first estimating motion vectors using unilateral (jointing forward and backward) and bilateral motion estimation. Then motion vectors are combined based on adaptive th...
[ "Hanieh Naderi", "Mohammad Rahmati" ]
[ "eess.IV", "cs.AI", "cs.CV", "cs.MM" ]
[ "Engineering", "Computer Science" ]
2022-03-05T00:00:00
https://arxiv.org/abs/2203.03621
https://arxiv.org/pdf/2203.03621v1
2203.03621
10.48550/arXiv.2203.03621
0
0
false
null
arXiv.org
0
4776b7a754e2e630469baf90d21410631a79bfc25ece756c1cad1e4081149ee8
[ "arxiv", "semantic_scholar" ]
Model Reprogramming: Resource-Efficient Cross-Domain Machine Learning
In data-rich domains such as vision, language, and speech, deep learning prevails to deliver high-performance task-specific models and can even learn general task-agnostic representations for efficient finetuning to downstream tasks. However, deep learning in resource-limited domains still faces multiple challenges inc...
[ "Pin-Yu Chen" ]
[ "cs.LG", "cs.AI" ]
[ "Computer Science" ]
2022-02-22T00:00:00
https://arxiv.org/abs/2202.10629
https://arxiv.org/pdf/2202.10629v4
2202.10629
10.1609/aaai.v38i20.30267
87
4
true
https://github.com/IBM/model-reprogramming
AAAI Conference on Artificial Intelligence
0.4861
b7c17ad48ab990be9f90f51ebb8835b89e8b79dcda5a32e0120b067c43e87103
[ "arxiv", "semantic_scholar" ]
FILM: Frame Interpolation for Large Motion
We present a frame interpolation algorithm that synthesizes multiple intermediate frames from two input images with large in-between motion. Recent methods use multiple networks to estimate optical flow or depth and a separate network dedicated to frame synthesis. This is often complex and requires scarce optical flow ...
[ "Fitsum Reda", "Janne Kontkanen", "Eric Tabellion", "Deqing Sun", "Caroline Pantofaru", "Brian Curless" ]
[ "cs.CV" ]
[ "Computer Science" ]
2022-02-10T00:00:00
https://arxiv.org/abs/2202.04901
https://arxiv.org/pdf/2202.04901v4
2202.04901
10.1007/978-3-031-20071-7_15
238
47
true
https://github.com/google-research/frame-interpolation
European Conference on Computer Vision
0.8406
ffee7bd98a6878025ec36e2fb397100cbcd6a9f86f689b16b752ad52a5147de7
[ "arxiv", "semantic_scholar" ]
Global-Reasoned Multi-Task Learning Model for Surgical Scene Understanding
Global and local relational reasoning enable scene understanding models to perform human-like scene analysis and understanding. Scene understanding enables better semantic segmentation and object-to-object interaction detection. In the medical domain, a robust surgical scene understanding model allows the automation of...
[ "Lalithkumar Seenivasan", "Sai Mitheran", "Mobarakol Islam", "Hongliang Ren" ]
[ "eess.IV", "cs.RO" ]
[ "Engineering", "Computer Science" ]
2022-01-28T00:00:00
https://arxiv.org/abs/2201.11957
https://arxiv.org/pdf/2201.11957v1
2201.11957
10.1109/LRA.2022.3146544
46
0
true
https://github.com/lalithjets/Global-reasoned-multi-task-model
IEEE Robotics and Automation Letters
0.418
aa0ffa9a8de37e85c387720b4480db74a94c91e5ffee730533ebcd63fa765b61
[ "arxiv", "semantic_scholar" ]
Arithmetic Monodromy Groups of Dynamical Belyi maps
We consider a large family of dynamical Belyi maps of arbitrary degree and study the arithmetic monodromy groups attached to the iterates of such maps. Building on the results of Bouw-Ejder-Karemaker on the geometric monodromy groups of these maps, we show that the quotient of the arithmetic monodromy group by the geom...
[ "Ozlem Ejder" ]
[ "math.NT" ]
[ "Mathematics" ]
2022-01-22T00:00:00
https://arxiv.org/abs/2201.09005
https://arxiv.org/pdf/2201.09005v1
2201.09005
null
5
0
false
null
null
0.1945
f94c08bf8846279809568f12e503970be7f6415d35dbb03100e3a47f56a06e39
[ "arxiv", "semantic_scholar" ]
Arithmetic geometric model for the renormalisation of irrationally indifferent attractors
In this paper we build a geometric model for the renormalisation of irrationally indifferent fixed points. The geometric model incorporates the fine arithmetic properties of the rotation number at the fixed point. Using this model for the renormalisation, we build a topological model for the dynamics of a holomorphic m...
[ "Davoud Cheraghi" ]
[ "math.DS", "math.FA" ]
[ "Mathematics", "Physics" ]
2021-12-29T00:00:00
https://arxiv.org/abs/2112.14557
https://arxiv.org/pdf/2112.14557v4
2112.14557
10.1088/1361-6544/ad0279
2
1
false
null
Nonlinearity
0.1505
4cbd23bb997c6caeb87eb7f799117f9ac5662a2a3d8c76dd6682534e2da4ed00
[ "arxiv", "semantic_scholar" ]
Modeling the debonding process of osseointegrated implants due to coupled adhesion and friction
Cementless implants have become widely used for total hip replacement surgery. The long-term stability of these implants is achieved by bone growing around and into the porous surface of the implant, a process called osseointegration. However, debonding of the bone-implant interface can still occur due to aseptic impla...
[ "Katharina Immel", "Vu-Hieu Nguyen", "Guillaume Haiat", "Roger A. Sauer" ]
[ "physics.med-ph", "cs.CE" ]
[ "Computer Science", "Physics", "Medicine" ]
2021-12-13T00:00:00
https://arxiv.org/abs/2112.06793
https://arxiv.org/pdf/2112.06793v2
2112.06793
10.1007/s10237-022-01637-7
7
0
false
null
Biomechanics and Modeling in Mechanobiology
0.2258
f07c63b47d071dd7944334bf04e7189603d597538a3a9eb11805e9f6bf410087
[ "arxiv", "semantic_scholar" ]
Merging Models with Fisher-Weighted Averaging
Averaging the parameters of models that have the same architecture and initialization can provide a means of combining their respective capabilities. In this paper, we take the perspective that this "merging" operation can be seen as choosing parameters that approximately maximize the joint likelihood of the posteriors...
[ "Michael Matena", "Colin Raffel" ]
[ "cs.LG" ]
[ "Computer Science" ]
2021-11-18T00:00:00
https://arxiv.org/abs/2111.09832
https://arxiv.org/pdf/2111.09832v2
2111.09832
10.52202/068431-1287
635
84
false
null
Neural Information Processing Systems
0.9647
5378acbe20da0b8fdbdbc06899f488c9ffece815ea92f3a1302c0cff00935ef8
[ "arxiv", "semantic_scholar" ]
Worst case expansions of complete theories
Given a complete theory $T$ and a subset $Y \subseteq X^k$, we precisely determine the {\em worst case complexity}, with respect to further monadic expansions, of an expansion $(M,Y)$ by $Y$ of a model $M$ of $T$ with universe $X$. In particular, although by definition monadically stable/NIP theories are robust under a...
[ "Samuel Braunfeld", "Michael C. Laskowski" ]
[ "math.LO" ]
[ "Mathematics" ]
2021-07-22T00:00:00
https://arxiv.org/abs/2107.10920
https://arxiv.org/pdf/2107.10920v2
2107.10920
10.2140/mt.2022.1.15
5
1
false
null
Model Th. 1 (2022) 15-30
0.1945
470fd8abb2e9af7fe04279721da06a545a1cfe1a19cf483531d0d4d21ecbede1
[ "arxiv", "semantic_scholar" ]
Self-training with noisy student model and semi-supervised loss function for dcase 2021 challenge task 4
This report proposes a polyphonic sound event detection (SED) method for the DCASE 2021 Challenge Task 4. The proposed SED model consists of two stages: a mean-teacher model for providing target labels regarding weakly labeled or unlabeled data and a self-training-based noisy student model for predicting strong labels ...
[ "Nam Kyun Kim", "Hong Kook Kim" ]
[ "cs.SD", "cs.LG", "eess.AS" ]
[ "Computer Science", "Engineering" ]
2021-07-06T00:00:00
https://arxiv.org/abs/2107.02569
https://arxiv.org/pdf/2107.02569v1
2107.02569
null
14
1
false
null
arXiv.org
0.294
02b0495b4938a34a44958de4b5cf6d7f058f0fbdb74917c25a086958dc7673ca
[ "arxiv", "semantic_scholar" ]
Interpolation and Model Checking for Nonlinear Arithmetic
We present a new model-based interpolation procedure for satisfiability modulo theories (SMT). The procedure uses a new mode of interaction with the SMT solver that we call solving modulo a model. This either extends a given partial model into a full model for a set of assertions or returns an explanation (a model inte...
[ "Dejan Jovanović", "Bruno Dutertre" ]
[ "cs.LO", "cs.PL", "cs.SC" ]
[ "Computer Science" ]
2021-06-08T00:00:00
https://arxiv.org/abs/2106.04340
https://arxiv.org/pdf/2106.04340v1
2106.04340
10.1007/978-3-030-81688-9_13
4
1
false
null
International Conference on Computer Aided Verification
0.1747
36de292de3329dc39d77c23462908988c91a54df4b1b1d57bf3109a6fac7f19f
[ "arxiv", "semantic_scholar" ]
Musical Prosody-Driven Emotion Classification: Interpreting Vocalists Portrayal of Emotions Through Machine Learning
The task of classifying emotions within a musical track has received widespread attention within the Music Information Retrieval (MIR) community. Music emotion recognition has traditionally relied on the use of acoustic features, verbal features, and metadata-based filtering. The role of musical prosody remains under-e...
[ "Nicholas Farris", "Brian Model", "Richard Savery", "Gil Weinberg" ]
[ "cs.SD", "cs.LG", "eess.AS" ]
[ "Computer Science", "Engineering" ]
2021-06-04T00:00:00
https://arxiv.org/abs/2106.02556
https://arxiv.org/pdf/2106.02556v2
2106.02556
null
1
0
false
null
arXiv.org
0.0753
b8321bd11df614dc10182956d8c23e9d0889bc68415c7e40987e515027383f2a
[ "arxiv", "semantic_scholar" ]
Towards physically consistent data-driven weather forecasting: Integrating data assimilation with equivariance-preserving deep spatial transformers
There is growing interest in data-driven weather prediction (DDWP), for example using convolutional neural networks such as U-NETs that are trained on data from models or reanalysis. Here, we propose 3 components to integrate with commonly used DDWP models in order to improve their physical consistency and forecast acc...
[ "Ashesh Chattopadhyay", "Mustafa Mustafa", "Pedram Hassanzadeh", "Eviatar Bach", "Karthik Kashinath" ]
[ "physics.ao-ph", "cs.AI", "cs.LG", "physics.comp-ph" ]
[ "Computer Science", "Physics" ]
2021-03-16T00:00:00
https://arxiv.org/abs/2103.09360
https://arxiv.org/pdf/2103.09360v1
2103.09360
10.5194/GMD-2021-71
42
3
false
null
arXiv.org
0.4084
695ab4feac12ddcff9312b75eca1cbbb9e27fbb960762249f2afe8577c651ecf
[ "arxiv", "semantic_scholar" ]
Analysis of Interpolation based Image In-painting Approaches
Interpolation and internal painting are one of the basic approaches in image internal painting, which is used to eliminate undesirable parts that occur in digital images or to enhance faulty parts. This study was designed to compare the interpolation algorithms used in image in-painting in the literature. Errors and no...
[ "Mustafa Zor", "Erkan Bostanci", "Mehmet Serdar Guzel", "Erinc Karatas" ]
[ "cs.CV", "eess.IV" ]
[ "Computer Science", "Engineering" ]
2021-02-12T00:00:00
https://arxiv.org/abs/2102.06564
https://arxiv.org/pdf/2102.06564v1
2102.06564
10.1201/9781003221333-8
1
0
false
null
null
0.0753