arXiv:2606. 31422v1 Announce Type: new Abstract: Long-horizon language agents do not only choose actions; they carry a private model of the world from one decision to the next.
arXiv:2606. 31422v2 Announce Type: replace Abstract: Language agents acting over long horizons must maintain beliefs about tool states, object locations, graph edges, and subgoal dependencies.
GAVEL is a framework that uses an explicit graph world model to verify and repair long‑horizon plans generated by large language models (LLMs). The graph encodes object relations, action pre‑conditions and effects, and probabilistic beliefs about unobserved object locations, allowing the system to predict action outcomes, detect violations, and repair them before execution. In experiments on BEHAVIOR‑1K, GAVEL boosts single‑task success from 41.2 % to 91.8 % and multi‑task success from 19.9 % to 92.6 %, while also reducing travel distance by about 5.4 % compared with a static variant.
arXiv:2609.00455v1 Announce Type: new Abstract: Large language models (LLMs) are being used as policies for autonomous decision-making and planning in many domains. Despite their strong reasoning cap...
The paper introduces NavMCP, a scaffolding framework that couples vision‑language models (VLMs) with navigation foundation models (NFMs) to enable long‑horizon physical‑world agents. NavMCP orchestrates three communication channels—intent, observation, and memory—to allow the VLM to decide what evidence to seek and the NFM to ground semantic sub‑goals into closed‑loop navigation, without retraining either model. The approach achieves state‑of‑the‑art results on several embodied question‑answering benchmarks and significantly outperforms episodic interfaces on the Unitree Go2 robot as task horizons lengthen.
arXiv:2606. 03685v1 Announce Type: cross Abstract: Supervised fine-tuning (SFT) improves end-to-end classical planning in large language models (LLMs), but do these models also learn to represent and reason about the planning problems they are solving?
Beyond Memory: Harnessing Long-Horizon Agents with Explicit Belief States introduces PoS, an inference-time framework that builds and maintains explicit belief states to guide large language model agents. Each belief state combines an estimate of the current world with unresolved task requirements, making clear what the agent still needs to learn and accomplish. PoS validates consistency, monitors task progress to detect Belief Trapping, and tailors recovery to the trapping pattern and unresolved requirements, achieving top performance across four benchmarks with all three LLM backbones.
arXiv:2606. 14574v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed as planners for autonomous agents in household environments.
The paper introduces Consistent Plan-Act (ConPAct), a method that addresses coordination failures between high-level planners and low-level actors in long-horizon agentic tasks. By prompting both agents to produce structured state assertions and programmatically detecting contradictions, the authors identify a systematic planner-actor state mismatch. ConPAct feeds these detected contradictions back to both agents, fine‑tunes them on consistent interactions, and achieves notable performance gains, such as raising MiniGrid success rates from 38.6% to 54.4% with GPT‑5.6‑sol/terra.
Visual Language Models (VLMs) excel at describing visible scene content but struggle to reason about dynamic multi-agent interactions, where action semantics depend on coordinated roles and spatial-te...
CivBench is an open‑source benchmark that evaluates language‑model agents in the long‑horizon, tool‑mediated game Civilization VI using the Model Context Protocol (MCP). Each episode lasts over 300 turns, generating thousands of tool calls across a 76‑tool action space, and includes a narration layer that translates visual game state into structured text. The study characterises agent behaviour across four model families, introducing Proactive Monitoring Rate (PMR) and RAG@10 as interface‑level metrics, and finds that agents often under‑monitor strategic state and fail to execute near‑term commitments despite tool access and explicit guidance.
The paper introduces DoublesEval, a diagnostic framework that uses professional doubles badminton to test visual‑language models’ ability to reason about dynamic multi‑agent interactions. It decomposes rallies into key moments and evaluates models across four dimensions—atomic recognition, intra‑segment composite understanding, cross‑segment causal reasoning, and high‑level tactical abstraction—highlighting specific reasoning failures. The authors also propose TacticCheck, a lightweight consistency checker that improves performance without retraining the models, yet significant gaps remain in tactical reasoning.
arXiv:2607. 02686v1 Announce Type: new Abstract: Reinforcement learning agents operating under partial observability must act on incomplete information, making them natural candidates for guidance from small language models (SLMs) that carry broad reasoning priors.