10 papers · 1 filter
From Reasoning to Action: Consequence-Grounded Alignment for Vision-Language-Action Models
Yixue Zhang, Wenke Xia, Kun Wu +6
Reasoning informs decision-making by connecting intended actions with their expected consequences. Recent vision-language-action (VLA) models incorporate embodied chains of thought…
Event2Reward: Learning Robotic Rewards from Progress Relations
Junnan Nie, Zitian Fan, Jiachen Zhang +14
Learning task progress from robot demonstrations provides a practical way to construct dense shaping rewards for robotic reinforcement learning. A common approach uses normalized d…
CASE: Cost-Aware Stopping for Efficient Long-Video Agents
Yiming Du, Chenghao Liu, Zhiyuan Liu +4
Long-video agents can actively gather question-relevant evidence, but they typically leave a central decision implicit: when has the agent seen enough to answer? We propose CASE, a…
A Theory of Reliable Self-Evolution for Agent Harnesses
Qianshu Cai, Yonggang Zhang, Jun Nie +6
In harness self-evolution, agents modify their own prompts, code, tools, and orchestration while keeping the underlying language model fixed. Recent work has shown that agents can…
EvolveNet: Collaborative Harness Evolution for Agent Self-Improvement
Jun Nie, Yonggang Zhang, Qianshu Cai +3
An LLM agent can acquire persistent improvements by evolving its harness---the executable program governing model calls, tool use, verification, and recovery---without updating mod…
Generated Images Are Easier to Forget: A Machine Unlearning Perspective for Synthetic Image Detection
Jun Nie, Yonggang Zhang, Tongliang Liu +3
Robust detection of generated images is critical to counter the misuse of generative models. Existing methods primarily depend on learning from human-annotated training datasets, l…
PACE: Phase-Aware Chunk Execution for Robot Policies with Action Chunking
Junnan Nie, Jiayi Li, Jiachen Zhang +5
Action-chunking robot policies predict sequences of future actions, but deployment requires choosing how many actions to execute before replanning. The execution horizon trades off…
DRNOISE: Benchmarking Deep Research Agents in Misleading Evidence Environments
Jun Nie, Zhiqin Yang, Zhenheng Tang +4
Deep research agents increasingly operate over the open web, where relevant records coexist with redundant summaries, outdated reports, and misleading documents. Existing evaluatio…
TTHE: Test-Time Harness Evolution
Jun Nie, Yonggang Zhang, Jun Song +5
The behavior of an LLM agent is determined not only by the underlying model, but also by its harness: the executable program that constructs context, invokes tools, verifies interm…
What Frozen VLAs Already Know About Success: A Probing Study of Value-Like Structure in Foundation Robot Policies
Jiachen Zhang, Junnan Nie, Junyi Lao +4
Vision--language--action (VLA) policies are trained to imitate actions; their loss never asks them to estimate reward, progress, or future success. Their frozen representations nev…