Zhijiang Guo

Beyond Pass@ 1: Self-Play with Variational Problem Synthesis Sustains RLVR

SWINGARENA: Adversarial Programming Arena for Long-context GitHub Issue Solving

Decomposing Policy Optimization into Proxy Objective and Knowledge Distillation for Long-context Reasoning

ACE: Attribution-Controlled Knowledge Editing for Multi-hop Factual Recall

Developmental Federated Tuning: A Cognitive-Inspired Paradigm for Efficient LLM Adaptation

Prune to Fit: Enabling Federated Fine-Tuning within Edge Memory Budgets

Elastic Mixture of Rank-Wise Experts for Knowledge Reuse in Federated Fine-Tuning

When Silence Is Golden: Can LLMs Learn to Abstain in Temporal QA and Beyond?

Less is More: Resource-Efficient Low-Rank Adaptation

Latent-DPO: Direction-Aware Preference Optimization for Reasoning Alignment

TL;DR: Too Long, Do Re-weighting for Efficient LLM Reasoning Compression

Depth-Breadth Synergy in RLVR: Unlocking LLM Reasoning Gains with Adaptive Exploration

Critique to Verify: Accurate and Honest Test-Time Scaling with RL-Trained Verifiers

TreeRPO: Tree Relative Policy Optimization

Determine-Then-Ensemble: Necessity of Top-k Union for Large Language Model Ensembling

Activation-Guided Consensus Merging for Large Language Models

OptiBench Meets ReSocratic: Measure and Improve LLMs for Optimization Modeling

FormalAlign: Automated Alignment Evaluation for Autoformalization

Atom of Thoughts for Markov LLM Test-Time Scaling

Measuring, Evaluating and Improving Logical Consistency in Large Language Models

UNComp: Uncertainty-Aware Long-context Compressor for Efficient Large Language Model Inference

Aligning with Logic: Measuring, Evaluating and Improving Logical Preference Consistency in Large Language Models

Process-Driven Autoformalization in Lean 4

CtrlA: Adaptive Retrieval-Augmented Generation via Inherent Control

EffiCoder: Enhancing Code Generation in Large Language Models through Efficiency-Aware Fine-tuning

MHPP: Exploring the Capabilities and Limitations of Language Models Beyond Basic Code Generation

Effi-Code: Unleashing Code Efficiency in Language Models