Min Zhang

Revisiting Long-context Modeling from Context Denoising Perspective

Supervised Fine-Tuning or Contrastive Learning? Towards Better Multimodal LLM Reranking

Is On-Policy Data always the Best Choice for Direct Preference Optimization-Based LM Alignment?

REA-RL: Reflection-Aware Online Reinforcement Learning for Efficient Reasoning

Evaluating and Improving Cultural Awareness of Reward Models for LLM Alignment

$Re^{2}$: Unlocking LLM Reasoning via Reinforcement Learning with Re-solving

KaLM-Embedding-V2: Superior Training Techniques and Data Inspire A Versatile Embedding Model

LongRM: Revealing and Unlocking the Context Boundary of Reward Modeling

Multi-objective Large Language Model Alignment with Hierarchical Experts

LycheeDecode: Accelerating Long-Context LLM Inference via Hybrid-Head Sparse Decoding

HiPO: Self-Hint Policy Optimization for RLVR

Skill Weaving: Efficient Self-Improvement of LLMs via Modular Skillpacks

Echoes as Anchors: Probabilistic Costs and Attention Refocusing in LLM Reasoning

FAPO: Flawed-Aware Policy Optimization for Efficient and Reliable Reasoning

Evaluating and Steering Modality Preference in Multimodal Large Language Model

CaliDrop: KV Cache Compression with Query-based Calibration

LoCoT2V-Bench: A Benchmark for Long-Form and Complex Text-to-Video Generation

SplitLoRA: Balancing Stability and Plasticity in Continual Learning Through Gradient Space Splitting

HURST: Learning Heterogeneity-Adaptive Urban Foundation Models for Spatiotemporal Prediction via Self-Partitional Mixture-of-Spatial-Experts

Language Bias in LVLMs: From In-Depth Analysis to Simple and Effective Mitigation

Uni-DPO: A Unified Paradigm for Dynamic Preference Optimization of LLMs

Improving Code Translation Correctness and Efficiency with Multi-Perspective Exploration and Difference-Aware Selection

Knowledge Fusion of Large Language Models via Modular SkillPacks

Where Matters More Than What: Decoding-aligned KV Cache Compression via Position-aware Pseudo-queries

The Hidden Link Between RLHF and Contrastive Learning

VIPO-R1: Cultivating Video Reasoning in MLLMs via Verifier-Guided Iterative Policy Optimization

VC-Bench: Pioneering the Video Connecting Benchmark with a Dataset and Evaluation Metrics

Unifying Reinforcement Learning and Distillation via Distribution Matching for Video Generation

Backdoors in RLVR: Jailbreak Backdoors in LLMs From Verifiable Reward

Training-Free Self-Scheduling for Efficient LLM Inference Serving

LongFlow: Efficient KV Cache Compression for Reasoning Models

Vocabulary Fixation Reveals Visual Attention Sink for Hallucination Mitigation in LVLMs

Self-Healing: Recovering Pruned Large Reasoning Models via Reinforcement Learning

BatonVoice: An Operationalist Framework for Enhancing Controllable Speech Synthesis with Linguistic Intelligence from LLMs

From Vulnerability to Defense: Understanding and Mitigating MASK-Based Attacks in dLLMs

Controllable and Interpretable Multi-Value Alignment For Large Language Model

From Bias to Balance: Exploring and Mitigating Spatial Bias in LVLMs

GRM-Omni: Generative Omni-modality Reward Modeling via Meta Reward Learning

XBOUND: Exploring Capability Boundaries of Device-Control Agents at the State Level

Introspective Adversarial Learning: Autonomous and Continual Preference Learning for LLM Alignment

Adaptive Mixing of Non-Invariant Information for Generalized Diffusion Policy

SCAN: Self-Denoising Monte Carlo Annotation for Robust Process Reward Learning

Exploring the Translation Mechanism of Large Language Models

Phased Training for LLM-powered Text Retrieval Models Beyond Data Scaling

Revealing and Mitigating Over-Attention in Knowledge Editing

Thinking in Character: Advancing Role-Playing Agents with Role-Aware Reasoning

Function-to-Style Guidance of LLMs for Code Translation

DelTA: An Online Document-Level Translation Agent Based on Multi-Level Memory

MASTER: Enhancing Large Language Model via Multi-Agent Simulated Teaching

Learning to Watermark: A Selective Watermarking Framework for Large Language Models via Multi-Objective Optimization

IDInit: A Universal and Stable Initialization Method for Neural Network Training

Reflection on Knowledge Graph for Large Language Models Reasoning

DISRetrieval: Harnessing Discourse Structure for Long Document Retrieval

LOGO --- Long cOntext aliGnment via efficient preference Optimization

Unleashing Reasoning Capability of LLMs via Scalable Question Synthesis from Scratch

Beware of Calibration Data for Pruning Large Language Models

Multi-modality Expansion and Retention for LLMs through Parameter Merging and Decoupling

LOGO --- Long cOntext aliGnment via efficient preference Optimization

InfCycle: Learning to Use Tools via Inference Compute and Cycle Consistency

DynamicKV: Task-Aware Adaptive KV Cache Compression for Long Context LLMs

SinkQ: Accurate 2-bit KV Cache Quantization with Dynamic Sink Tracking

Path Selection Makes BERT-family Good Generators

L-CiteEval: Do Long-Context Models Truly Leverage Context for Responding?

CASD: Enhancing Generation Accuracy via Context-Aware Speculative Decoding

Evolving Virtual World with Delta-Engine