1 citations · 1 across the 21 of their papers we have counts for
26 papers
ImageRSI: Verification-Driven Recursive Self-Improvement for Agentic Image Generation
Shuo Han, Zhilin Wang, Peng Ling +5
Complex text-to-image requests remain difficult to satisfy reliably. Verifying and repairing an image can fix an individual output, but a successful repair does not necessarily car…
A Lightweight Velocity Correction Framework for Improving Instruction Following in Few-Step Text-to-Image Generation
Yating Gao, Zhenlong Zhu, Wen Zihao +6
Few-step text-to-image generators produce high-quality images with few network forward passes, yet struggle with complex prompts involving multiple subjects, interactions, spatial…
Teacher-Student Gaps Are Not Enough: Outcome-Guided On-Policy Distillation for Multi-Turn Autonomous Agents
Tong Zhang, Zhou Liu, Yihao Liu +8
On-policy distillation (OPD) trains a student on its own trajectories with dense teacher supervision. Recent work on OPD for multi-turn autonomous agents often treats large teacher…
Token-Disentangled Latent Test-Time Scaling for Vision-Language Reasoning
Hao-Xuan Ma, Yihao Liu, Yutao Sun +8
Latent test-time scaling improves reasoning by refining hidden states during inference, but existing methods typically apply a single scalar reward to all editable latent tokens. F…
SMOPD: Multi-Reward Reinforcement Learning via Specialize-and-Merge Online Policy Distillation
Wen Wang, Jiahua Bao, Tu Yongsiqi +8
We aim to improve model performance in multi-reward reinforcement learning training process. Existing Group reward-Decoupled Normalization Policy Optimization (GDPO) has mitigated…
Skill Self-Play: Pushing the Frontier of LLM Capability with Co-Evolving Skills
Siyuan Huang, Pengyu Cheng, Haotian Liu +10
LLM training is shifting from manual design and annotation to interaction-driven self-evolution. However, existing self-evolutionary methods face a fundamental dilemma between task…
From Passive Retrieval to Active Memory Navigation: Learning to Use Memory as a Structured Action Space
Yue Xu, Yutao Sun, Yihao Liu +7
Long-term user memory is essential for personalized conversational agents, yet many memory systems still expose memory through passive retrieval interfaces, making the model a cons…
Dynamo: Dynamic Skill-Tool Evolution for Vision-Language Agents
Yutao Sun, Yanting Miao, Hao-Xuan Ma +8
Improving vision-language models (VLMs) on visual reasoning typically requires retraining or hand-designed prompts and tools. We present Dynamo, a training-free framework that adap…
PolicyAlign: Direct Policy-Based Safety Alignment for Large Language Models
Chang Wu, Junfeng Fang, Houcheng Jiang +5
Safety alignment of large language models (LLMs) typically depends on high-quality supervision data, such as safe demonstrations or preference pairs. However, in real-world deploym…
A Survey of Full-Duplex Spoken Dialogue Systems: Architectural Hierarchy, Interaction Ontology, and Decision State Machine
Jingyu Lu, Yuhan Wang, Jianming Luo +15
More than a dozen spoken dialogue systems have recently claimed to be "full-duplex," yet the term has been used to describe substantially different capabilities. Existing surveys c…
Learning from Your Own Mistakes: Constructing Learnable Micro-Reflective Trajectories for Self-Distillation
Zhilin Huang, Hang Gao, Ziqiang Dong +6
Self-distillation improves reasoning in large language models by using the model's own rollouts as training signal, typically through implicit logit-level alignment that minimizes…
GDPO: Mitigating Multi-Reward Conflicts via Group-Dynamic reward-Decoupled Policy Optimization
Haotian Liu, Yihao Liu, Jingwei Ni +11
As LLMs advance, post-training reinforcement learning (RL) increasingly relies on multi-dimensional rewards to cultivate comprehensive capabilities. This shift demands new algorith…
Skill-RM: Unifying Heterogeneous Evaluation Criteria via Agent Skill
Tao Chen, Gangwei Jiang, Pengyu Cheng +10
Reward models (RMs) provide critical feedback signals for LLM post-training, notably in reinforced fine-tuning (RFT) and reinforcement learning (RL) pipelines. However, current rew…
Fill the GAP: A Granular Alignment Paradigm for Visual Reasoning in Multimodal Large Language Models
Yanting Miao, Yutao Sun, Dexin Wang +8
Visual latent reasoning lets a multimodal large language model (MLLM) create intermediate visual evidence as continuous tokens, avoiding external tools or image generators. However…
Revisiting Reinforcement Learning with Verifiable Rewards from a Contrastive Perspective
Feng Zhang, Xinhong Ma, Ziqiang Dong +5
Group Relative Policy Optimization (GRPO) is one of the most widely adopted RLVR algorithms for post-training large language models on reasoning tasks. We first show that GRPO admi…
MARCH: Multi-Agent Reinforced Self-Check for LLM Hallucination
Zhuo Li, Yupeng Zhang, Pengyu Cheng +8
Hallucination remains a critical bottleneck for large language models (LLMs), undermining their reliability in real-world applications, especially in Retrieval-Augmented Generation…
Rationale Matters: Learning Transferable Rubrics via Proxy-Guided Critique for VLM Reward Models
Weijie Qiu, Dai Guan, Junxin Wang +6
Generative reward models (GRMs) for vision-language models (VLMs) often evaluate outputs via a three-stage pipeline: rubric generation, criterion-based scoring, and a final verdict…
CLIPO: Contrastive Learning in Policy Optimization Generalizes RLVR
Sijia Cui, Pengyu Cheng, Jiajun Song +6
Reinforcement Learning with Verifiable Rewards (RLVR) has significantly advanced the reasoning capacity of Large Language Models (LLMs). However, RLVR solely relies on final answer…
ATP-Bench: Towards Agentic Tool Planning for MLLM Interleaved Generation
Yinuo Liu, Zi Qian, Heng Zhou +7
Interleaved text-and-image generation represents a significant frontier for Multimodal Large Language Models (MLLMs), offering a more intuitive way to convey complex information. C…
Trace2Skill: Distill Trajectory-Local Lessons into Transferable Agent Skills
Jingwei Ni, Yihao Liu, Xinpeng Liu +7
Large Language Model (LLM) agents increasingly rely on domain-specific skills, yet manually authoring such skills does not scale, and skills generated purely from parametric knowle…
Grounding the Score: Explicit Visual Premise Verification for Reliable Vision-Language Process Reward Models
Junxin Wang, Dai Guan, Weijie Qiu +7
Vision-language process reward models (VL-PRMs) are increasingly used to score intermediate reasoning steps and rerank candidates under test-time scaling. However, they often funct…
SiameseNorm: Breaking the Barrier to Reconciling Pre/Post-Norm
Tianyu Li, Dongchen Han, Zixuan Cao +7
The long-standing tension between Pre- and Post-Norm remains an open problem in Transformer architecture, reflecting a fundamental trade-off between training stability and represen…
Eliminating Inductive Bias in Reward Models with Information-Theoretic Guidance
Zhuo Li, Pengyu Cheng, Zhechao Yu +7
Reward models (RMs) are essential in reinforcement learning from human feedback (RLHF) to align large language models (LLMs) with human values. However, RM training data is commonl…
Bringing Stability to Diffusion: Decomposing and Reducing Variance of Training Masked Diffusion Models
Mengni Jia, Mengyu Zhou, Yihao Liu +2
Masked diffusion models (MDMs) are a promising alternative to autoregressive models (ARMs), but they suffer from inherently much higher training variance. High variance leads to no…