20 results for “long-horizon tasks”
CS papers onlyHybrid search: Keyword + semantic, ranked by combined score.ⓘ
Want pure semantic search? Try claim verification →
Kewei Xu, Xiaoben Lu, Shuofei Qiao, Zihan Ding +3 more
The paper introduces LongDS, a new benchmark for long-horizon, multi-turn data analysis, demonstrating that current AI agents struggle significantly with maintaining and updating complex analytical st…
Xuancheng Zhu, Yang Yue, Shuaibing Wan, Zihan Dou +3 more
The paper introduces TaskWeave, a hierarchical agentic framework that successfully simulates long-horizon organizational dynamics by treating coordination as a memory-centered problem, demonstrating t…
Yuan Xiong, Linji Hao, Shizhu He, Yequan Wang +1 more
The paper proposes Janus, a framework for long-horizon agent safety using multi-agent simulation and CoAA-RL, which trains guards to anticipate risks and make safety judgments based on both observed p…
This paper recompiles time horizons for AI capabilities using splines and item-response theory to improve their interpretability.
A model-based planner is presented that minimizes immediate cost and aggregated expected future cost across all robots in a task planning scenario with robots sharing a persistent environment.
This paper studies the complexity of Value-based Task Selection (VTS) in real-time systems.
Boxiu Li, Zimo Wen, Yijia Fan, Junxiang Lei +22 more
Argus is a persistent, self-evolving runtime system that achieves better long-horizon reasoning results than Direct Copilot in seven GPT-5.5 benchmark arenas, with improvements in solve-input tokens,…
Evaluates the capabilities of language model agents in a simulated real-world task.
The paper introduces RSPM, a memory management technique for agents that improves performance by pruning and forgetting unnecessary information.
The paper introduces GTA, a scalable framework for generating realistic, multi-hop web-agent tasks with dense, executable trajectories, addressing the current lack of process-level supervision in web…
RACE-Sched is an asynchronous agentic framework that successfully integrates low-latency, real-time scheduling decisions with advanced, long-horizon reasoning provided by Large Language Models.
The paper introduces new benchmarks for complex asynchronous planning and demonstrates that general constraint satisfaction formalizers (like CP-SAT) significantly outperform direct LLM planning or tr…
This paper introduces a controlled environment to study multi-turn long-horizon planning ability acquisition, shaping, and integration in foundation model agents.
The paper introduces GONDOR, a memory-efficient extension of Greedy Best-First Search (GBFS) that enables search continuation under strict memory constraints by periodically compressing the search tre…
The paper introduces a new construct, await_delete, to allow safe value reuse in futures and presents a novel future-based algorithm for block-wise inversion of dense, symmetric matrices.
This paper investigates various methods for encoding factored tasks, a compact planning representation, into propositional logic for use with SAT solvers, analyzing the impact of encoding choices and…