paper-reading

https://github.com/nuaa-nlp/paper-reading/blob/main/README.md

Guideline:

• paper reading讲解的时候要深入浅出，确保自己看懂了，再用通俗的话讲出来。关键是把文章工作讲清楚，motivation，方法部分，实验是否支撑，该工作的优点和缺点，对你个人工作的启发。最重要的是后面两部分，需要你自己对工作批判性的阅读。
• 分享的同学务必提前告知大家分享的论文，并在分享前update paper信息及slides到 nuaa-nlp/paper-reading；新人权限开通请联系pjli。
• 参与者希望都能够提前把分享的paper进行相关背景的了解，积极提出问题及参与讨论。

next reading

2025/11/21

Speakers	Papers	Slides	Others
Yixin Bu	ICR Probe: Tracking Hidden State Dynamics for Reliable Hallucination Detection in LLMs	[slides]	-

2025/11/14

Speakers	Papers	Slides	Others
Shuo Feng	Bootstrapping Language-Guided Navigation Learning with Self-Refining Data Flywheel	[slides]	-
-	NavRAG: Generating User Demand Instructions for Embodied Navigation through Retrieval-Augmented LLM	-	-

2025/5/23

Speakers	Papers	Slides	Others
Hongkai Zheng	Parallel Scaling Law for Language Models	[slides]	-

2025/05/09

Speakers	Papers	Slides	Others
Xinyan Shi	ReSearch: Learning to Reason with Search for LLMsvia Reinforcement Learning	[slides]	-
-	R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learning	-	-
-	Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning

2025/4/25

Speakers	Papers	Slides	Others
Bo Zhang	Rethinking Transformers Pre-training for Multi-Spectral Satellite Imagery	[slides]	-
-	SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagery	-	-
-	Masked Autoencoders Are Scalable Vision Learners	-	-

2025/4/18

Speakers	Papers	Slides	Others
Teng Lin	SePer: MEASURE RETRIEVAL UTILITY THROUGH THE LENS OF SEMANTIC PERPLEXITY REDUCTION	[slides]	-

2025/3/28

Speakers	Papers	Slides	Others
Guanyun Zou	REDEEP: DETECTING HALLUCINATION IN RETRIEVAL-AUGMENTED GENERATION VIA MECHANISTIC INTERPRETABILITY	[slides]	-

2025/3/21

Speakers	Papers	Slides	Others
Yongpeng Zhang	Explanations of Deep Language Models Explain Language Representations in the Brain	[slides]	-

2025/01/03

Speakers	Papers	Slides	Others
Shuo Feng	PRET: Planning with Directed Fidelity Trajectory for Vision and Language Navigation	[slides]	-
-	Exploring Temporal Concurrency for Video-Language Representation Learning	-	-
-	Language modeling via stochastic processes	-	-
Guanyun Zou	DRAGIN: Dynamic Retrieval Augmented Generation based on the Information Needs of Large Language Models	[slides]	-

2024/12/06

Speakers	Papers	Slides	Others
Feiyan Zhai	BEWARE OF CALIBRATION DATA FOR PRUNING LARGE LANGUAGE MODELS	[slides]	-
-	DDK: Distilling Domain Knowledge for Efficient Large Language Models	-	-

2024/11/22

Speakers	Papers	Slides	Others
Runze Xia	TOPOLM: BRAIN-LIKE SPATIO-FUNCTIONAL ORGANIZATION IN A TOPOGRAPHIC LANGUAGE MODEL	[slides]	-

2024/11/15

Speakers	Papers	Slides	Others
HongKai Zheng	Neural Discrete Representation Learning	[slides]	-
-	Addressing Representation Collapse in Vector Quantized Models with One Linear Layer	-	-
-	Finite Scalar Quantization: VQ-VAE Made Simple	-	-

2024/10/25

Speakers	Papers	Slides	Others
Zexuan Li	ERA-CoT: Improving Chain-of-Thought through Entity Relationship Analysis	[slides]	-

2024/10/25

Speakers	Papers	Slides	Others
Congchi Yin	Fast Inference from Transformers via Speculative Decoding	[slides]	-

2024/10/18

Speakers	Papers	Slides	Others
Yixin Bu	SEMANTIC UNCERTAINTY: LINGUISTIC INVARIANCES FOR UNCERTAINTY ESTIMATION IN NATURAL LANGUAGE GENERATION	[slides]	-

2024/10/11

Speakers	Papers	Slides	Others
Yongpeng Zhang	Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection	[slides]	-

2024/09/27

Speakers	Papers	Slides	Others
Xinyan Shi	Physics of Language Models: Part 2.1, Grade-School Math and the Hidden Reasoning Process	[slides]	-

2024/09/20

Speakers	Papers	Slides	Others
Yangsong Lan	Dataset Distillation	[slides]	-
-	Dataset Condensation with Gradient Matching	-	-
-	Dataset Condensation with Distribution Matching	-	-

2024/09/13

Speakers	Papers	Slides	Others
Bo Zhang	Fortify the Shortest Stave in Attention: Enhancing Context Awareness of Large Language Models for Effective Tool-Use	[slides]	-

2024/05/31

Speakers	Papers	Slides	Others
Shuo Feng	NOIR: Neural Signal Operated Intelligent Robots for Everyday Activities	[slides]	-

2024/04/19

Speakers	Papers	Slides	Others
Xinyan Shi	Physics of Language Models: Part 3.1, Knowledge Storage and Extraction	[slides]	-

2024/04/07

Speakers	Papers	Slides	Others
Xuanfan Ni	Mamba: Linear-Time Sequence Modeling with Selective State Spaces	[slides]	-

2024/03/15

Speakers	Papers	Slides	Others
Bo Zhang	Trial and Error: Exploration-Based Trajectory Optimization for LLM Agents	[slides]	-

2024/03/08

Speakers	Papers	Slides	Others
Feiyan Zhai	Dynamic Confidence-Aware Multi-Modal Emotion Recognition	[slides]	-

2024/01/17

Speakers	Papers	Slides	Others
Shuo Feng	Structure-Encoding Auxiliary Tasks for Improved Visual Representation	[slides]	-

2024/1/10

Speakers	Papers	Slides	Others
Yixin Bu	Meaning without reference in large language models	[slides]	-

2023/12/27

Speakers	Papers	Slides	Others
Yongpeng Zhang	Evidence of a predictive coding hierarchy in the human brain listening to speech	[slides]	-

2023/12/20

Speakers	Papers	Slides	Others
Renzhi Wang	Label Words are Anchors: An Information Flow Perspective for Understanding In-Context Learning	[slides]	-

2023/12/13

Speakers	Papers	Slides	Others
Zexuan Li	LLMaAA: Making Large Language Models as Active Annotators	[slides]	-
-	LabelPrompt: Effective Prompt-based Learning for Relation Classification	-	-

2023/11/29

Speakers	Papers	Slides	Others
Yangsong Lan	Go Wider Instead of Deeper	[slides]	-
-	Towards Being Parameter-Efficient: A Stratified Sparsely Activated Transformer with Dynamic Capacity	-	-
-	Mixture-of-Experts with Expert Choice Routing	-	-
-	Brainformers: Trading Simplicity for Efficiency	-	-

2023/11/22

Speakers	Papers	Slides	Others
Haiyang Ou	Ultra-fine Entity Typing with Indirect Supervision from Natural Language Inference	[slides]	-
-	PromptNER : Prompting For FewShot Named Entity Recognition	-	-

2023/11/15

Speakers	Papers	Slides	Others
Yang Cao	On Faithfulness and Factuality in Abstractive Summarization	[slides]	-
-	GPTEval: NLG Evaluation using GPT-4 with Better Human Alignment	-	-

2023/11/08

Speakers	Papers	Slides	Others
Xi Wang	ChatHaruhi: Reviving Anime Character in Reality via Large Language Model	[slides]	-
-	RoleLLM: Benchmarking, Eliciting, and Enhancing Role-Playing Abilities of Large Language Models	-	-

2023/11/03

Speakers	Papers	Slides	Others
Congchi Yin	Disentangling Syntax and Semantics in the Brain with Deep Networks	[slides]	-

2023/10/25

Speakers	Papers	Slides	Others
Ruoqing Zhao	Finding-Aware Anatomical Tokens for Chest X-Ray Automated Reporting	[slides]	-
-	Knowledge-enhanced Visual-Language Pre-training on Chest Radiology Images	-	-

2023/10/18

Speakers	Papers	Slides	Others
Runze Xia	MindGPT: Interpreting What You See with Non-invasive Brain Recordings	[slides]	-
-	BrainSCUBA: Fine-Grained Natural Language Captions of Visual Cortex Selectivity	-	-

2023/10/11

Speakers	Papers	Slides	Others
Renzhi Wang	Editing Large Language Models: Problems, Methods, and Opportunities	[slides]	-
-	EasyEdit: An Easy-to-use Knowledge Editing Framework for Large Language Models	-	-
Xinyan Shi	MemPrompt: Memory-assisted Prompt Editing with User Feedback	[slides]	-
-	Aging with GRACE: Lifelong Model Editing with Discrete Key-Value Adapters	-	-

2023/09/27

Speakers	Papers	Slides	Others
Fan Yuan	Chain-of-Verification Reduces Hallucination in Large Language Models	[slides]	-
-	Evaluating Object Hallucination in Large Vision-Language Models	-	-

Speakers	Papers	Slides	Others
Xuanfan NI	DOLA: DECODING BY CONTRASTING LAYERS IMPROVES FACTUALITY IN LARGE LANGUAGE MODELS	[slides]	-
-	Knowledge Sanitization of Large Language Models	-	-

2023/09/25

Speakers	Papers	Slides	Others
Feiyan Zhai	Go Wider Instead of Deeper	[slides]	-
-	Pushing Mixture of Experts to the Limit:Extremely Parameter Efficient MoE for Instruction Tuning	-	-
-	OUTRAGEOUSLY LARGE NEURAL NETWORKS:THE SPARSELY-GATED MIXTURE-OF-EXPERTS LAYER	-	-

Speakers	Papers	Slides	Others
BoZhang	Generative Agents: Interactive Simulacra of Human Behavior	[slides]	-
-	CHATEVAL: TOWARDS BETTER LLM-BASED EVALUATORS THROUGH MULTI-AGENT DEBATE	-	-
-	BOLAA: Benchmarking and Orchestrating LLM-augmented Autonomous Agents	-	-
-	AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation Framework	-	-

2023/09/13

Speakers	Papers	Slides	Others
Yihong Liu	Highly accurate protein structure prediction with AlphaFold	[slides]	-
-	Robust deep learning based protein sequence design using ProteinMPNN	-	-
-	ProteinBERT: A universal deep-learning model of protein sequence and function	-	-
-	ProtGPT2 is a deep unsupervised language model for protein design	-	-
-	OntoProtein: Protein Pretraining With Gene Ontology Embedding	-	-

2023/09/13

Speakers	Papers	Slides	Others
Shuo Feng	NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language Models	[slides]	-
-	TEACh: Task-driven Embodied Agents that Chat	-	-

2023/06/09

Speakers	Papers	Slides	Others
Feiyan Zhai	FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness	[slides]	-

2023/05/05

Speakers	Papers	Slides	Others
Renzhi Wang	Generative Agents: Interactive Simulacra of Human Behavior	[slides]	-

2023/04/21

Speakers	Papers	Slides	Others
Xuanfan Ni	Deep reinforcement learning from human preferences	[slides]	-
-	Check Your Facts and Try Again: Improving Large Language Models with External Knowledge and Automated Feedback	-	-

2023/04/07

Speakers	Papers	Slides	Others
Xi Wang	Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?	[slides]	-
-	Larger language models do in-context learning differently	-	-

2023/03/23

Speakers	Papers	Slides	Others
Fan Yuan	Language Is Not All You Need: Aligning Perception with Language Models	[slides]	-
-	Visual ChatGPT: Talking, Drawing and Editing with Visual Foundation Models	-	-

2023/03/17

Speakers	Papers	Slides	Others
Yang Cao	Get To The Point: Summarization with Pointer-Generator Networks	[slides]	-
-	SimCLS: A Simple Framework for Contrastive Learning of Abstractive Summarization	-	-

2023/03/03

Speakers	Papers	Slides	Others
Congchi Yin	findings of ACL 2022 Logic-Driven Context Extension and Data Augmentation for Logical Reasoning of Text	[slides]	-
-	findings of ACL 2022 MERIt: Meta-Path Guided Contrastive Learning for Logical Reasoning	-	-

2022/12/06

Speakers	Papers	Slides	Others
Shuo Feng	Vision-and-Language Navigation: Interpreting visually-grounded navigation instructions in real environments	[slides]	-
-	VLNœBERT: A Recurrent Vision-and-Language BERT for Navigation	-	-
-	General Evaluation for Instruction Conditioned Navigation using Dynamic Time Warping	-	-
-	Learning Disentanglement with Decoupled Labels for Vision-Language Navigation	-	-
-	LOViS: Learning Orientation and Visual Signals for Vision and Language Navigation	-	-

2022/11/29

Speakers	Papers	Slides	Others
Renzhi Wang	Denoising Diffusion Probabilistic Models	[slides]	-
-	Improved Denoising Diffusion Probabilistic Models	-	-
-	Score-Based Generative Modeling through Stochastic Differential Equations	-	-
-	Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecasting	-	-
-	Diffusion Models Beat GANs on Image Synthesis	-	-
-	High-Resolution Image Synthesis With Latent Diffusion Models	-	-
-	Diffusion-LM Improves Controllable Text Generation	-	-
	Denoising Diffusion Implicit Models	-	-
-	DiffuSeq: Sequence to Sequence Text Generation with Diffusion Models	-	-
-	Structured Denoising Diffusion Models in Discrete State-Spaces	-	-
-	DiffusER: Discrete Diffusion via Edit-based Reconstruction	-	-
-	SSD-LM: Semi-autoregressive Simplex-based Diffusion Language Model for Text Generation and Modular Control	-	-

2022/11/15

Speakers	Papers	Slides	Others
Xuanfan Ni (NUAA)	Chain of Thought Prompting Elicits Reasoning in Large Language Models	[slides]	-
-	Self-Consistency Improves Chain of Thought Reasoning in Language Models	[slides]	-

2022/11/01

Speakers	Papers	Slides	Others
Fan Yuan	ICLR 2019 The neuro-symbolic concept learner: Interpreting scenes, words, and sentences from natural supervision	[slides]	-

2022/10/18

Speakers	Papers	Slides	Others
Feiyan Zhai	NeurIPS 2020 Denoising Diffusion Probabilistic Models	[slides]	-
-	Diffusion-LM Improves Controllable Text Generation	-	-

2022/10/11

Speakers	Papers	Slides	Others
Xi Wang	ICML 2020 Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation	[slides]	-

2022/09/27

Speakers	Papers	Slides	Others
Ruoqing Zhao	ICLR 2022 BEiT: BERT Pre-Training of Image Transformers	[slides]	-
-	BEiT v2: Masked Image Modeling with Vector-Quantized Visual Tokenizers	-	-
-	Image as a Foreign Language: BEiT Pretraining for All Vision and Vision-Language Tasks	-	-

2022/09/20

Speakers	Papers	Slides	Others
Yang Cao	NIPS 2017 Attention is all you need	[slides]	-

2022/09/13

Speakers	Papers	Slides	Others
Congchi Yin	ICLR 2021 An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale	[slides]	-
-	ICCV 2021 Swin Transformer: Hierarchical Vision Transformer using Shifted Windows	-	-

2022/04/01

Speakers	Papers	Slides	Others
Xuan Sheng	ICLR 2020 Heterofl: Computation and communication efficient federated learning for heterogeneous clients	[slides]	-
-	NeurIPS workshop 2019 FedMD: Heterogenous Federated Learning via Model Distillation	-	-

2022/03/25

Speakers	Papers	Slides	Others
Zhaoyang Han	CCS 2017 Practical Secure Aggregation for Privacy-Preserving Machine Learning	[slides]	-
-	MLSys 2019 TOWARDS FEDERATED LEARNING AT SCALE: SYSTEM DESIGN	[slides]	-

2021/12/22

Speakers	Papers	Slides	Others
Xi Wang(NUAA)	Chinese Spelling Check paper sharing	[slides]	-
-	SpellGCN: Incorporating Phonological and Visual Similarities into Language Models for Chinese Spelling Check	-	-
-	Correcting Chinese Spelling Errors with Phonetic Pre-training	-	-
-	PHMOSpell: Phonological and Morphological Knowledge Guided Chinese Spelling Check	-	-

2021/12/15

Speakers	Papers	Slides	Others
RuoQing Zhao(NUAA)	causal inference and medical report generation paper sharing	[slides]	-
-	TieNet: Text-Image Embedding Network for Common Thorax Disease Classification and Reporting in Chest X-rays	-	-
-	Causal Attention for Vision-Language Tasks	-	-

2021/12/02

Speakers	Papers	Slides	Others
Fan Yuan (NUAA)	multi-modal dialogue paper sharing	[slides]	-
-	Multi-Modal Open-Domain Dialogue	-	-
-	Multimodal Dialogue Response Generation	-	-
-	Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts	-	-

2021/12/02

Speakers	Papers	Slides	Others
Weibin Wu (NUAA)	A Brief Tutorial of Side Channel Attack	[slides]	-

2021/11/25

Speakers	Papers	Slides	Others
Zhixin Zhao (NUAA)	A Brief Tutorial of Blockchain	[slides]	-

2021/11/17

Speakers	Papers	Slides	Others
Xuanfan Ni (NUAA)	Prefix-Tuning: Optimizing Continuous Prompts for Generation	[slides]	-
-	A Character-Centric Neural Model for Automated Story Generation	[slides]	-
-	Attention Is All You Need	[slides]	-

2021/11/11

Speakers	Papers	Slides	Others
Zeyu Qin (CUHK-SZ)	A Brief Tutorial of Adversarial Machine Learning (After 2019)	[slides]	-

2021/10/28

Speakers	Papers	Slides	Others
Wenjie Zhou	ICDE 2021 Attacking Black-box Recommendations via Copying Cross-domain User Profiles	[slides]	-

2021/10/28

Speakers	Papers	Slides	Others
ZiHao Deng	CVPR 2021 MAZE: Data-Free Model Stealing Attack Using Zeroth-Order Gradient Estimation	[slides]	-

2021/10/21

Speakers	Papers	Slides	Others
Zhicheng Li	unpublish 2021 Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing.	[slides]	-

2021/09/30

Speakers	Papers	Slides	Others
Zikang Jin	CVPR 2019 Evading defenses to transferable adversarial examples by translation-invariant attacks.	[slides]	-
-	CVPR 2019 Feature space perturbations yield more transferable adversarial examples.	-	-

2021/09/16

Speakers	Papers	Slides	Others
Yundi Shi	EMNLP 2017 Adversarial Examples for Evaluating Reading Comprehension Systems.	[slides]	-
-	ACL 2019 Improving the Robustness of Question Answering Systems to Question Paraphrasing.	-	-

2021/09/02

Speakers	Papers	Slides	Others
Xuan Sheng	USENIX 2020 Hybrid Batch Attacks: Finding Black-box Adversarial Examples with Limited Queries	[slides]	-
-	CCS 2020 Gotta Catch'Em All: Using Honeypots to Catch Adversarial Attacks on Neural Networks	-	-

2021/08/18

Speakers	Papers	Slides	Others
Changchun Yin	ICLR 2014 Intriguing properties of neural networks	[slides]	-
-	ICLR 2015 Explaining and Harnessing Adversarial Examples	-	-
-	IJCAI 2018 Generating Adversarial Examples with Adversarial Networks	-	-
-	TEC 2019 One Pixel Attack for Fooling Deep Neural Networks	-	-

2021/07/28

Speakers	Papers	Slides	Others
Zhaoyang Han	USENIX 2020 TEXTSHIELD: Robust Text Classification Based on Multimodal Embedding and Neural Machine Translation	[slides]	-
-	NDSS 2019 TextBugger: Generating Adversarial Text Against Real-world Applications	-	-