It is argued that grounding explanations in a formalized representation enables interactive affordances beyond what static text supports, and is instantiated for mathematical proof comprehension with explorable theorems, a system that uses LLMs to translate a theorem and its written proof into Lean, a programming langu...
Hita Kambhamettu, Will Crichton, S. Welleck et al.· 0 citations
Computer-use agents have demonstrated strong capabilities in interpreting screens and executing keyboard and mouse actions. However, games, with their visually rich worlds and diverse mechanics, remain challenging and offer a demanding next step toward general digital control. Current game-agent learning pipelines ofte...
Ming-Yu Ouyang, Yi-Ran Wang, Si-Yuan Hu et al.· 0 citations
We propose DeltaStreamer, a novel agentic framework for streaming video understanding inspired by how humans observe, remember, and investigate. Streaming video understanding requires a system to retain evidence before knowing which details future queries will need. Existing approaches often obscure event transitions a...
Bo-Yuan Sun, De-Shui Miao, Wen-Zhao Gao et al.· 0 citations
Cooperative success rests in part on implicit communication. Partners converge on conventions: regularities that give an action meaning beyond its direct effect. Yet conventions are usually observed only indirectly: a score drop when partners are swapped shows that a convention existed, but not what it was or where it...
Makoto Fukushima, Tim Puphal· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
This work introduces task-state governance, an interaction paradigm that represents a task as an editable contract linking desired outcomes, facts that must be preserved, and observed application state in TaskVM, which projects live state into editable task controls.
Wen-Kui Yang, Yi-Ming Hao, Kai Zhou et al.· 0 citations
Existing methods induce various forms of knowledge from agent traces, from sample workflows to programmatic skills. We propose ARISE (Agent Routine Induction from Successful Experience), which identifies recurring procedures in agent traces and synthesizes them into a reusable library of agent subroutines. Across five...
Zhihan Lei, Jiarui Yan, Joshua Momo et al.· 0 citations
Many recent document embedding models are trained on document-as-image representations, embedding rendered pages as images rather than the underlying source. Meanwhile, existing benchmarks for scientific document retrieval, such as ArXivQA and ViDoRe, treat documents as images of pages, implicitly favoring such represe...
Ghazal Khalighinejad, Raghuveer Thirukovalluru, Alexander H. Oh et al.· 0 citations
It is found that external rewards drive most of the accuracy gains, while preserving more coherent reasoning when rewarded for reasoning tokens, while preserving more coherent reasoning when rewarded for reasoning tokens through internal attention rewards.
There has been growing interest in building agents that can interact with digital platforms to execute meaningful enterprise tasks autonomously. Among the approaches explored are tool-augmented agents built on abstractions such as Model Context Protocol (MCP) and web agents that operate through graphical interfaces. Ye...
Large language models (LLMs) increasingly store user preferences in persistent memory to support personalization across interactions. However, in third-party communication settings governed by social and institutional norms, some user preferences may be inappropriate to apply. We introduce BenchPreS, which evaluates wh...
Sangyeon Yoon, SunKyoung Kim, Hyesoo Hong et al.· arXiv.org· 8 citations· ⚡2
Draft-Thinking is proposed, a lightweight adaptation framework that learns a compact, self-contained working derivation that directly supports the final answer in a long chain-of-thought reasoning model.
Jie Cao, Tianwei Lin, Zhen-Xuan Fan et al.· arXiv.org· 3 citations· ⚡1
Crafting effective reward signals remains a central challenge in reinforcement learning (RL) for complex reasoning. Existing methods for automated reward optimization search with a fixed proposer which never learns from its costly trials. We present Dual-Evolving Reinforcement Learning (DERL), a dual-training framework...
Sitao Cheng, Tianle Li, Xuhan Huang et al.· 0 citations
