arXiv Computation and Language
TreeFlex: Agents Navigating Their Own Search Trees
Read the original on arXiv Computation and Language →The Flow has not summarised this story yet — read it at arXiv Computation and Language.
Related stories
Hugging Face Trending Papers
5d ago
TreeFlex: Agents Navigating Their Own Search Trees
A ReAct-style agent accumulates observations, intermediate results, and failed attempts in a single context. As this context grows, stale and off-path information consumes its limited capacity and obs...
arXiv AI
Jun 11
TreeSeeker: Tree-Structured Trial, Error, and Return in Deep Search
arXiv:2606. 11662v1 Announce Type: new Abstract: Deep search requires agents to answer complex questions through multi-step web search, browsing, evidence comparison, and synthesis.
Hugging Face Trending Papers
Jun 10
TreeSeeker: Tree-Structured Trial, Error, and Return in Deep Search
Deep search requires agents to answer complex questions through multi-step web search, browsing, evidence comparison, and synthesis. A central challenge is deciding how to search when several directions look plausible but only some will later lead to reliable evidence.
Hugging Face Trending Papers
Jul 9
GATS: Graph-Augmented Tree Search with Layered World Models for Efficient Agent Planning
Large Language Model (LLM) agents have shown promise in multi-step planning tasks, but existing approaches like LATS (Language Agent Tree Search) and ReAct rely heavily on LLM inference during planning, leading to high computational costs and stochastic behavior. We present \textbf{GATS} (Graph-Augmented Tree Search), a planning framework that combines systematic UCB1-based tree search with a layered world model to eliminate LLM calls during inference while achieving superior planning performance.
arXiv AI
4d ago
Probe-and-Resume: When and How to Evaluate Agent Execution
The paper investigates the use of test‑time search, specifically Monte Carlo tree search (MCTS), to improve agent performance in terminal‑based computer‑use environments. It finds that conventional MCTS does not significantly boost task success due to limited critic information and poor intermediate state evaluation. The authors propose a new strategy called Probe‑and‑Resume, where a critic is invoked only when the agent believes it has finished, and it is given read, write, and execute access to a forked copy of the environment to probe hidden state. This approach raises success rates by 8.4 percentage points over standard MCTS and nearly matches a more expensive Codex CLI agent.
arXiv Machine Learning
5d ago
AtomSearch: Evidence-Grounded Tree Search for Agent Skill Optimization
arXiv:2610.19951v1 Announce Type: cross Abstract: Agent skills adapt frozen models through editable external instructions. Revising several rules at once makes changes difficult to inspect, while jud...