Bo Han

Decoupling the Class Label and the Target Concept in Machine Unlearning

On the Thinking-Language Modeling Gap in Large Language Models

LLM Unlearning with LLM Beliefs

Memorize to Forget: Machine Unlearning without Gradient Ascent via Model Extrapolation

Towards Understanding Valuable Preference Data for Large Language Model Alignment

Co-rewarding: Stable Self-supervised RL for Eliciting Reasoning in Large Language Models

AEGIS: Adversarial Target-Guided Retention-Data-Free Robust Concept Erasure from Diffusion Models

EEPO: Exploration-Enhanced Policy Optimization via Sample-Then-Forget

Concept Concentration for Faithful Representation Intervention

Task-Aware Data Selection via Proxy-Label Enhanced Distribution Matching for LLM Finetuning

Detecting Generated Images via Machine Unlearning

Reasoned Safety Alignment: Ensuring Jailbreak Defense via Answer-Then-Check

Beyond Raw Detection Scores: Markov-Informed Calibration for Boosting Machine-Generated Text Detection

On the Fragility of Latent Knowledge: Layer-wise Influence under Unlearning in Large Language Model

AdLift: Lifting Adversarial Perturbations to Safeguard 3D Gaussian Splatting Assets Against Instruction-Driven Editing

Weak-to-Strong GraphRAG: Aligning Weak Retrievers with Large Language Models for Graph-based Retrieval Augmented Generation

Reference-guided Policy Optimization for Molecular Optimization via LLM Reasoning

Landscape of Thoughts: Visualizing the Reasoning Process of Large Language Models

TriSpec: Ternary Speculative Decoding via Lightweight Proxy Verification

Machine-Generated Text Detection Requires Fewer Machine-Human Mixed Texts

CoDaPO: Confidence and Difficulty-Adaptive Policy Optimization for Language Models

FRAG: Filtering Noise Using Snippet-Level Query Relevance

Bilateral Information-aware Test-time Adaptation for Vision-Language Models

Sequence Length Matters in Data Scheduling for Accelerating Language Model Pretraining

MaGA: Machine-Guided Amnesiac Unlearning through Target Feature Disentanglement

Parameter-wise Weighted Model Editing for Efficient and Retentive LLM Unlearning

Towards Scalable Oversight with Collaborative Multi-Agent Debate in Error Detection

Symmetry-Guaranteed Prediction of High-Order Tensor Properties for Crystalline Materials via Irreducible Decomposition

What Is Preference Optimization Doing, How and Why?

R²AG: Learning to Reason, Retrieve, and Self-Evolve through a Multi-Branch Retrieval Tree

A Closer Look at In-context Learning of LLMs in Simple Classification Tasks

Multi-Agent Debate with Memory Masking

JailbreakLoRA: Your Downloaded LoRA from Sharing Platforms might be Unsafe

OracleKV: Oracle Guidance for Question-Independent KV Cache Eviction

ENTP: Enhancing Low-Quality SFT Data via Neural-Symbolic Text Purge-Mix

Lifelong Unlearning for Multimodal Large Language Models

Lang-PINN: From Language to Physics-Informed Neural Networks via a Multi-Agent Framework

Next-Token Gradient Sensitivity Probing for LLM Hallucination Detection

LeGIT: LLM Guided Intervention Targeting for Online Causal Discovery

Observe-Then-Think: Learning to Elicit Multimodal Understanding by Decoupling Perception and Reasoning

RewardFlow: Propagating Reward in the State Graphs of Agentic Learning with LLMs

Forgetting: A New Mechanism Towards Better Large Language Model Fine-tuning

Combining-on-Graph: Structured Reasoning of LLMs with Knowledge Graph

VeriBench: End-to-End Formal Verification Benchmark for AI Code Generation in Lean 4

Color Blindness Test Images as Seen by Large Vision-Language Models

Detecting LLM Hallucinations via Nonlinear Manifold Separation

Generative Model Inversion Through the Lens of the Manifold Hypothesis

Advancing Machine-Generated Text Detection from an Easy to Hard Supervision Perspective

Adaptive Localization of Knowledge Negation for Continual LLM Unlearning

Instance-dependent Early Stopping

Physics-Driven Spatiotemporal Modeling for AI-Generated Video Detection

Learning to Instruct for Visual Instruction Tuning

Fast and Accurate Blind Flexible Docking

Practical Kernel Selection for Kernel-based Conditional Independence Test

Epistemic Uncertainty for Generated Image Detection

Detecting Generated Images by Fitting Natural Image Distributions

Unlocker: Disentangle the Deadlock of Learning between Label-noisy and Long-tailed Data

A Robust Method to Discover Causal or Anticausal Relation

COSDA: Counterfactual-based Susceptibility Risk Framework for Open-Set Domain Adaptation

FedGPS: Statistical Rectification Against Data Heterogeneity in Federated Learning

A Lens into Interpretable Transformer Mistakes via Semantic Dependency

From Debate to Equilibrium: Belief‑Driven Multi‑Agent LLM Reasoning via Bayesian Nash Equilibrium

GRU: Mitigating the Trade-off between Unlearning and Retention for LLMs

Understanding and Enhancing the Transferability of Jailbreaking Attacks

Towards Out-of-Modal Generalization without Instance-level Modal Correspondence

Atomas: Hierarchical Adaptive Alignment on Molecule-Text for Unified Molecule Understanding and Generation

Rethinking LLM Unlearning Objectives: A Gradient Perspective and Go Beyond

Hot-pluggable Federated Learning: Bridging General and Personalized FL via Dynamic Selection

Noisy Test-Time Adaptation in Vision-Language Models

Decoupling the Class Label and the Target Concept in Machine Unlearning

Overfitting: An Unexpected Asset in AI‐Generated Image Detection

Breaking the Detection-Generalization Paradox on Out-Of-Distribution Data

Towards Effective Evaluations and Comparisons for LLM Unlearning Methods

Enhancing Sample Selection Against Label Noise by Cutting Mislabeled Easy Examples

Towards Efficient and Scalable Multi-agent Reasoning via Bayesian Nash Equilibrium

Learning without Isolation: Pathway Protection for Continual Learning

One-step Image-function Generation via Consistency Training

Detecting Discrepancies Between Generated and Natural Images Using Uncertainty

Noise Prompt Learning: Learning the Winning Tickets for Diffusion Sampling

On the Language of Thoughts in Large Language Models

Exploring Criteria of Loss Reweighting to Enhance LLM Unlearning

From Passive to Active Reasoning: Can Large Language Models Ask the Right Questions under Incomplete Information?

Exploring the Discriminative Capability of LLMs in In-context Learning

Strengthening Federated Learning: Surrogate Data-Guided Aggregation for Robust Backdoor Defense

Practical Kernel Learning for Kernel-based Conditional Independent Test

Single-Step Diffusion Model-Based Generative Model Inversion Attacks

Rectifying Gradient-based OOD Detection via Fisher Information Matrix

Understanding Mistakes in Transformers through Token-level Semantic Dependencies

When Data-Free Knowledge Distillation Meets Non-Transferable Teacher: Escaping Out-of-Distribution Trap is All You Need

Consistency Verification for Detecting AI-Generated Images

Nested Gloss Makes Large Language Models Lost

Can Large Language Models Help Experimental Design for Causal Discovery?

Do We Really Need Parameter-Isolation to Protect Task Knowledge?

Safeguarding System Prompts: A Surrogate-Based Defense Against Injection Attacks