17 papers
SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD
Dongfang Li, Xiaodong Luo, Ruoyu Sun +62
Full-parameter post-training of trillion-parameter-scale MoE models introduces substantial system-level challenges for large-scale distributed training, including severe memory pre…
From Profiling to Synthesis: Benchmarking Implicit Behavioral Alignment in Personalized LLM Agents
Jiajia Song, Bobo Li, Haiwen Yi +6
Large Language Models have enabled increasingly capable autonomous agents, yet personalization remains critical for making such agents practically useful. Recent benchmarks have be…
Agent Reinforcement Learning via Pivotal-Aware Self-Feedback Retry
Weiyang Guo, Zesheng Shi, Longhui Zhang +3
Large language model (LLM) agents have shown strong decision-making capabilities in long-horizon interactive tasks, yet they still struggle to effectively leverage failed trajector…
SuCo: Sufficiency-guided Continuous Adaptive Reasoning
Jiahao Wang, Bingyu Liang, Chenhao Hu +5
Despite remarkable performance on complex tasks, Large Reasoning Models (LRMs) often generate excessively long Chain-of-Thoughts (CoT), inflating computational costs even for simpl…
OGER: A Robust Offline-Guided Exploration Reward for Hybrid Reinforcement Learning
Xinyu Ma, Mingzhou Xu, Xuebo Liu +4
Recent advancements in Reinforcement Learning with Verifiable Rewards (RLVR) have significantly improved Large Language Model (LLM) reasoning, yet models often struggle to explore…
Mitigating Context-Memory Conflicts in LLMs through Dynamic Cognitive Reconciliation Decoding
Yigeng Zhou, Wu Li, Yifan Lu +6
Large language models accumulate extensive parametric knowledge through pre-training. However, knowledge conflicts occur when outdated or incorrect parametric knowledge conflicts w…
Taming Actor-Observer Asymmetry in Agents via Dialectical Alignment
Bobo Li, Rui Wu, Zibo Ji +5
Large Language Model agents have rapidly evolved from static text generators into dynamic systems capable of executing complex autonomous workflows. To enhance reliability, multi-a…
D-QRELO: Training- and Data-Free Delta Compression for Large Language Models via Quantization and Residual Low-Rank Approximation
Junlin Li, Shuangyong Song, Guodong Du +6
Supervised Fine-Tuning (SFT) accelerates taskspecific large language models (LLMs) development, but the resulting proliferation of finetuned models incurs substantial memory overhe…
Step-GRPO: Internalizing Dynamic Early Exit for Efficient Reasoning
Benteng Chen, Weida Wang, Shufei Zhang +2
Large reasoning models that use long chain-of-thought excel at problem-solving yet waste compute on redundant checks. Curbing this overthinking is hard: training-time length penalt…
MathAgent: Adversarial Evolution of Constraint Graphs for Mathematical Reasoning Data Synthesis
Zixiong Yu, Jun Rao, Guhan Chen +5
Synthesizing high-quality mathematical reasoning data without human priors remains a significant challenge. Current approaches typically rely on seed data mutation or simple prompt…
Backdoors in RLVR: Jailbreak Backdoors in LLMs From Verifiable Reward
Weiyang Guo, Zesheng Shi, Zeen Zhu +3
Reinforcement Learning with Verifiable Rewards (RLVR) is an emerging paradigm that significantly boosts a Large Language Model's (LLM's) reasoning abilities on complex logical task…
E3-TIR: Enhanced Experience Exploitation for Tool-Integrated Reasoning
Weiyang Guo, Zesheng Shi, Liye Zhao +5
While Large Language Models (LLMs) have demonstrated significant potential in Tool-Integrated Reasoning (TIR), existing training paradigms face significant limitations: Zero-RL suf…
MASRubric: Auditing Information Flow in Multi-Agent Systems with Failure-Distilled Pitfall Rubrics
Yutong Wang, Siyuan Xiong, Xuebo Liu +4
While multi-agent systems (MAS) excel at complex reasoning, they are vulnerable to errors that intermediate agents introduce and downstream agents build upon. Auditing intermediate…
Mining or Synthesis? Rethinking Exploration Efficiency in Iterative Alignment of Mathematical Reasoning
Jun Rao, Zixiong Yu, Xuebo Liu +6
Iterative Direct Preference Optimization (DPO) has emerged as a widely used paradigm for aligning Large Language Models on reasoning tasks. Existing approaches typically rely on Be…
Tailoring Diagnostic Modeling to Individual Learners: Personalized Distractor Generation via MCTS-Guided Reasoning Reconstruction
Tao Wu, Jingyuan Chen, Wang Lin +6
Distractors-incorrect yet plausible answer choices in multiple-choice questions (MCQs)-are vital in educational assessments, as they help identify student misconceptions by present…
REA-RL: Reflection-Aware Online Reinforcement Learning for Efficient Reasoning
Hexuan Deng, Wenxiang Jiao, Xuebo Liu +2
Large Reasoning Models (LRMs) demonstrate strong performance in complex tasks but often face the challenge of overthinking, leading to substantially high inference costs. Existing…
Dynamic Sampling that Adapts: Self-Aware Iterative Data Persistent Optimization for Mathematical Reasoning
Jun Rao, Xuebo Liu, Hexuan Deng +5
In mathematical reasoning, data selection strategies predominantly rely on static, externally defined metrics, which fail to adapt to the evolving capabilities of models during tra…