Nanyun Peng

Energy-Regularized Sequential Model Editing on Hyperspheres

Multilingual Routing in Mixture-of-Experts

Steering MoE LLMs via Expert (De)Activation

FronTalk: Benchmarking Front-End Development as Conversational Code Generation with Multi-Modal Feedback

MM-PoisonRAG: Disrupting Multimodal RAG with Local and Global Knowledge Poisoning Attacks

LLM-REVal: Can We Trust LLM Reviewers Yet?

DialectGen: Benchmarking and Improving Dialect Robustness in Multimodal Generation

ARES: Multimodal Adaptive Reasoning via Difficulty-Aware Token-Level Entropy Shaping

Decoupling Task-Solving and Output Formatting in LLM Generation

SPARK: Stepwise Process-Aware Rewards for Reference-Free Reinforcement Learning

TemMed-Bench: Evaluating Temporal Medical Image Reasoning in Vision-Language Models

MMChat: A Multi-turn Multi-Modal Conversational Benchmark Grounded in Real World User Behaviors

MMPersuade: A Dataset and Evaluation Framework for Multimodal Persuasion

Do "New Snow Tablets" Contain Snow? Large Language Models Over-Rely on Names to Identify Ingredients of Chinese Drugs

On the Loss of Context Awareness in General Instruction Fine-tuning

PARTONOMY: Large Multimodal Models with Part-Level Visual Understanding

3DLLM-Mem: Long-Term Spatial-Temporal Memory for Embodied 3D Large Language Model

Contrastive Visual Data Augmentation

VaPR - Vision-language Preference alignment for Reasoning

OpenVLThinker: Complex Vision-Language Reasoning via Iterative SFT-RL Cycles

AI Debate Aids Assessment of Controversial Claims

InsNeXt: Training Scalable Insertion-based Language Models from Scratch

Unlocking Exocentric Video-Language Data for Egocentric Video Representation Learning

MRAG-Bench: Vision-Centric Evaluation for Retrieval-Augmented Multimodal Models

Scaling Probabilistic Circuits via Monarch Matrices

On the loss of context-awareness in general instruction finetuning

Model Extrapolation Expedites Alignment