Skip to content

Author

Pengzhan Sun

We have 6 of 7 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Ego2World: Compiling Egocentric Cooking Videos into Executable Worlds for Belief-State Planning

Ego2World is introduced, an executable benchmark that turns egocentric cooking videos into executable symbolic worlds governed by graph-transition rules and shows that persistent belief memory improves task completion while reducing repeated visual exploration, suggesting that belief maintenance should be a first-class...

Qin-Chuan Cheng, Zhan Gong, Peng-Zhan Sun et al. · 1 citation
#artificial intelligence Preprint Oct 2026

Ego2World: Compiling Egocentric Cooking Videos into Executable Worlds for Belief-State Planning

Egocentric videos capture how people carry out everyday activities, yet testing an agent requires evaluating the consequences of actions it chooses itself. We introduce Ego2World, a benchmark that turns annotated cooking activities into executable planning environments under partial observation. Its compiler links sour...

Qin-Chuan Cheng, Zhan Gong, Peng-Zhan Sun et al. · 1 citation
Preprint Oct 2026

From Reasoning Failures to Composable Video Spatial Intelligence

Spatial reasoning benchmarks evaluate vision-language models across diverse tasks, but task-level scores do not reveal which underlying capabilities account for success or failure. Each task requires recovering spatial evidence, representing geometry, and reasoning over it. We disentangle these capabilities by comparin...

Peng-Zhan Sun, Jun-Bin Xiao, Ramanathan Rajaraman et al. · 0 citations
Preprint Sep 2026

Aligning Thoughts with Answers: Probability Rewards to Tame Thinking Drift

Rita is proposed as a novel RL paradigm that performs consistently superior to the supervised finetuning approaches and vanilla RL-finetuned frameworks and adopts a difficulty-aware data filtering strategy that selects informative easy-to-medium samples for RL using rollout error rate and reward variance.

Peng-Zhan Sun, Shiu-hong Kao, Shi-Jie Li et al. · 1 citation
#artificial intelligence Preprint Oct 2026

Rethinking Probability-Based Reinforcement Learning From Posterior Concentration

Verifier-free reinforcement learning with probability-based rewards offers a promising way to train LLMs on general reasoning tasks where external verifiers are unavailable. Yet the reliability of these rewards, especially in long-horizon reasoning, remains underexplored. This work identifies a length-dependent failure...

Shiu-hong Kao, Yu-Bo Zhao, Zhen Tian et al. · 0 citations
Preprint Aug 2026

Dynamic Resolution Routing for Efficient Egocentric Grounding

This work proposes SmartRes, a framework that performs efficiency optimization in the pixel space via dynamic resolution routing and introduces a margin-regularized routing objective that increases foreground-background logit separation and improves foreground recall.

H. Sun, Wang-Bo Zhao, Fanyue Wei et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.