语言大模型 / LLM - arXivDaily 专题

2403.00876 2024-03-05 cs.CL cs.AI 62%

Word Order and World Knowledge

Qinghua Zhao, Vinit Ravishankar, Nicolas Garneau, Anders Søgaard

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.AI

2306.13292 2024-02-26 cs.LG cs.AI cs.CV 62%

Variance-Covariance Regularization Improves Representation Learning

Jiachen Zhu, Katrina Evtimova, Yubei Chen, Ravid Shwartz-Ziv, Yann LeCun

专题命中知识编辑与模型理解：pretraining(abstract)；分类 cs.AI、cs.LG

Comments 165 pages, 5 figures

2402.12560 2024-02-21 cs.CL cs.AI 62%

CausalGym: Benchmarking causal interpretability methods on linguistic tasks

Aryaman Arora, Dan Jurafsky, Christopher Potts

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.AI

Comments 9 pages main text, 26 pages total

2402.06957 2024-02-13 cs.CR cs.AI cs.CV cs.LG 62%

Architectural Neural Backdoors from First Principles

Harry Langford, Ilia Shumailov, Yiren Zhao, Robert Mullins, Nicolas Papernot

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.AI、cs.LG

2402.02541 2024-02-07 cs.CL cs.AI cs.CV 62%

Knowledge Generation for Zero-shot Knowledge-based VQA

Rui Cao, Jing Jiang

专题命中知识编辑与模型理解：LLM(abstract)；分类 cs.CL、cs.AI

Comments accepted as Findings in EACL 2023

2402.03137 2024-02-06 cs.CL cs.LG 62%

Sociolinguistically Informed Interpretability: A Case Study on Hinglish Emotion Classification

Kushal Tatariya, Heather Lent, Johannes Bjerva, Miryam de Lhoneux

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.LG

Comments 5 pages, Accepted to SIGTYP 2024 @ EACL

2402.01790 2024-02-06 cs.LG cs.AI 62%

An introduction to graphical tensor notation for mechanistic interpretability

Jordan K. Taylor

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.AI、cs.LG

Comments 30 pages, 75 figures

2401.11360 2024-01-23 cs.LG cs.AI cs.CE q-bio.BM 62%

PepHarmony: A Multi-View Contrastive Learning Framework for Integrated Sequence and Structure-Based Peptide Encoding

Ruochi Zhang, Haoran Wu, Chang Liu, Huaping Li, Yuqian Wu, Kewei Li, Yifan Wang, Yifan Deng, Jiahui Chen, Fengfeng Zhou, Xin Gao

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.AI、cs.LG

Comments 25 pages, 5 figures, 3 tables

2401.08619 2024-01-18 cs.LG cs.AI 62%

MATE-Pred: Multimodal Attention-based TCR-Epitope interaction Predictor

Etienne Goffinet, Raghvendra Mall, Ankita Singh, Rahul Kaushik, Filippo Castiglione

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.AI、cs.LG

Comments Patent pending: U.S. Provisional Application No. 63/603,952

2401.06709 2024-01-15 cs.CL cs.AI 62%

Reliability Analysis of Psychological Concept Extraction and Classification in User-penned Text

Muskan Garg, MSVPJ Sathvik, Amrit Chadha, Shaina Raza, Sunghwan Sohn

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.AI

2312.16104 2023-12-27 cs.CL cs.AI 62%

Dotless Representation of Arabic Text: Analysis and Modeling

Maged S. Al-Shaibani, Irfan Ahmad

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.AI

2305.14901 2023-12-27 cs.CL cs.LG 62%

Chain-of-Questions Training with Latent Answers for Robust Multistep Question Answering

Wang Zhu, Jesse Thomason, Robin Jia

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.LG

Comments Accepted by EMNLP 2023

2303.06854 2023-12-21 cs.CV cs.CL cs.CR cs.LG 62%

Robust Contrastive Language-Image Pre-training against Data Poisoning and Backdoor Attacks

Wenhan Yang, Jingdong Gao, Baharan Mirzasoleiman

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.LG

2312.06514 2023-12-12 cs.CL cs.AI 62%

Where exactly does contextualization in a PLM happen?

Soniya Vijayakumar, Tanja Bäumel, Simon Ostermann, Josef van Genabith

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.AI

Comments EMNLP 2023 BlackBloxNLP 2023 Workshop

2312.03721 2023-12-11 cs.CL cs.AI 62%

Exploring the Robustness of Model-Graded Evaluations and Automated Interpretability

Simon Lermen, Ondřej Kvapil

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.AI

2310.15020 2023-12-05 cs.RO cs.AI cs.CV cs.LG 62%

Invariance is Key to Generalization: Examining the Role of Representation in Sim-to-Real Transfer for Visual Navigation

Bo Ai, Zhanxin Wu, David Hsu

专题命中知识编辑与模型理解：foundation model(abstract)；分类 cs.AI、cs.LG

Comments 11 pages, accepted by the 18th International Symposium on Experimental Robotics (ISER 2023) and published within the Springer Proceedings in Advanced Robotics (SPAR)

2312.00751 2023-12-04 cs.CL cs.AI 62%

Mitigating Over-smoothing in Transformers via Regularized Nonlocal Functionals

Tam Nguyen, Tan M. Nguyen, Richard G. Baraniuk

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.AI

Comments 24 papes

2311.14583 2023-11-27 cs.CL cs.AI cs.IR 62%

GPT Struct Me: Probing GPT Models on Narrative Entity Extraction

Hugo Sousa, Nuno Guimarães, Alípio Jorge, Ricardo Campos

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.AI

2311.13258 2023-11-23 cs.CV cs.CL cs.LG 62%

ViStruct: Visual Structural Knowledge Extraction via Curriculum Guided Code-Vision Representation

Yangyi Chen, Xingyao Wang, Manling Li, Derek Hoiem, Heng Ji

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.LG

Comments Accepted to EMNLP 2023

2311.11701 2023-11-21 cs.IR cs.AI cs.CL cs.HC 62%

Control in Hybrid Chatbots

Thomas Rüdel, Jochen L. Leidner

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.AI

Comments 12 pages, 3 figures

2311.10127 2023-11-20 cs.AI cs.HC cs.LG 62%

Learning interactions to boost human creativity with bandits and GPT-4

Ara Vartanian, Xiaoxi Sun, Yun-Shiuan Chuang, Siddharth Suresh, Xiaojin Zhu, Timothy T. Rogers

专题命中知识编辑与模型理解：prompting(abstract)；分类 cs.AI、cs.LG

2311.08240 2023-11-15 cs.CL cs.AI 62%

Investigating the Encoding of Words in BERT's Neurons using Feature Textualization

Tanja Baeumel, Soniya Vijayakumar, Josef van Genabith, Guenter Neumann, Simon Ostermann

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.AI

Comments To be published in 'BlackboxNLP 2023: The 6th Workshop on Analysing and Interpreting Neural Networks for NLP'. Camera-ready version

2311.03287 2023-11-08 cs.LG cs.CL cs.CV 62%

Holistic Analysis of Hallucination in GPT-4V(ision): Bias and Interference Challenges

Chenhang Cui, Yiyang Zhou, Xinyu Yang, Shirley Wu, Linjun Zhang, James Zou, Huaxiu Yao

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.LG

2311.02171 2023-11-08 cs.LG cs.AI 62%

Emergence of Abstract State Representations in Embodied Sequence Modeling

Tian Yun, Zilai Zeng, Kunal Handa, Ashish V. Thapliyal, Bo Pang, Ellie Pavlick, Chen Sun

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.AI、cs.LG

Comments Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP 2023). Project webpage: https://abstract-state-seqmodel.github.io/

2311.02631 2023-11-07 cs.LG cs.AI 62%

A Critical Perceptual Pre-trained Model for Complex Trajectory Recovery

Dedong Li, Ziyue Li, Zhishuai Li, Lei Bai, Qingyuan Gong, Lijun Sun, Wolfgang Ketter, Rui Zhao

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.AI、cs.LG

Comments Accepted in ACM SIGSPATIAL 2023

2306.09539 2023-10-31 cs.CL cs.LG 62%

Block-State Transformers

Mahan Fathi, Jonathan Pilault, Orhan Firat, Christopher Pal, Pierre-Luc Bacon, Ross Goroshin

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.LG

Comments NeurIPS'23 - Thirty-seventh Conference on Neural Information Processing Systems

2310.17230 2023-10-27 cs.LG cs.CL 62%

Codebook Features: Sparse and Discrete Interpretability for Neural Networks

Alex Tamkin, Mohammad Taufeeque, Noah D. Goodman

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.LG

2310.16803 2023-10-26 cs.CL cs.LG 62%

Language Agnostic Code Embeddings

Saiteja Utpala, Alex Gu, Pin Yu Chen

专题命中知识编辑与模型理解：language model(abstract)；分类 cs.CL、cs.LG

2310.07582 2023-10-24 cs.LG cs.AI 62%

Linear Latent World Models in Simple Transformers: A Case Study on Othello-GPT

Dean S. Hazineh, Zechen Zhang, Jeffery Chiu

专题命中知识编辑与模型理解：foundation model(abstract)；分类 cs.AI、cs.LG

2310.11958 2023-10-19 cs.CL cs.LG 62%

Emptying the Ocean with a Spoon: Should We Edit Models?

Yuval Pinter, Michael Elhadad

专题命中知识编辑与模型理解：LLM(abstract)；分类 cs.CL、cs.LG

Comments Findings of ACL: EMNLP 2023

1. 知识编辑与模型理解 7583 篇

Word Order and World Knowledge

Variance-Covariance Regularization Improves Representation Learning

CausalGym: Benchmarking causal interpretability methods on linguistic tasks

Architectural Neural Backdoors from First Principles

Knowledge Generation for Zero-shot Knowledge-based VQA

Sociolinguistically Informed Interpretability: A Case Study on Hinglish Emotion Classification

An introduction to graphical tensor notation for mechanistic interpretability

PepHarmony: A Multi-View Contrastive Learning Framework for Integrated Sequence and Structure-Based Peptide Encoding

MATE-Pred: Multimodal Attention-based TCR-Epitope interaction Predictor

Reliability Analysis of Psychological Concept Extraction and Classification in User-penned Text

Dotless Representation of Arabic Text: Analysis and Modeling

Chain-of-Questions Training with Latent Answers for Robust Multistep Question Answering

Robust Contrastive Language-Image Pre-training against Data Poisoning and Backdoor Attacks

Where exactly does contextualization in a PLM happen?

Exploring the Robustness of Model-Graded Evaluations and Automated Interpretability

Invariance is Key to Generalization: Examining the Role of Representation in Sim-to-Real Transfer for Visual Navigation

Mitigating Over-smoothing in Transformers via Regularized Nonlocal Functionals

GPT Struct Me: Probing GPT Models on Narrative Entity Extraction

ViStruct: Visual Structural Knowledge Extraction via Curriculum Guided Code-Vision Representation

Control in Hybrid Chatbots

Learning interactions to boost human creativity with bandits and GPT-4

Investigating the Encoding of Words in BERT's Neurons using Feature Textualization

Holistic Analysis of Hallucination in GPT-4V(ision): Bias and Interference Challenges

Emergence of Abstract State Representations in Embodied Sequence Modeling

A Critical Perceptual Pre-trained Model for Complex Trajectory Recovery

Block-State Transformers

Codebook Features: Sparse and Discrete Interpretability for Neural Networks

Language Agnostic Code Embeddings

Linear Latent World Models in Simple Transformers: A Case Study on Othello-GPT

Emptying the Ocean with a Spoon: Should We Edit Models?