The paper introduces a fault‑tolerant budget conservation framework for distributed multi‑agent delegation, where budgets are represented as exclusive escrow credits that traverse a delegation DAG. It details how each branch converts credit into a reservation tied to lineage, epoch, and idempotency, persists a signed dispatch permit, and ensures that uncertain effects remain charged until settlement or retirement. The authors prove properties such as ownership partition, ledger conservation, and at‑most‑once settlement, and validate the mechanism through TLA+ checks, a JavaScript explorer, and crash‑injected SQLite experiments.
By Genliang Zhu, Chu Wang
The paper introduces a streaming-recall benchmark that separates retention and selection decisions for persistent agents. It shows that query‑aware selection boosts recall by 15.5 points when access is fixed, while mixed comparisons inflate gains due to changes in history access. The study finds that under bounded retention, failures stem from eviction rather than ranking errors, and that dense retrieval can outperform lexical retrieval on natural text.
By Juli Huang
Large language model agents depend on external harnesses to exchange information with their environment and to recover from execution errors, but recovery is typically evaluated only by overall task success, masking a key trade‑off. The authors treat recovery as a causal decision problem, comparing outcomes with and without recovery from the same execution state to separate rescue from harm and analyze how its value evolves over time. They propose the Causal Intervention Router (CIR), a lightweight policy that uses pre‑recovery information to decide when intervention is beneficial, achieving a 3‑point increase in success on long‑horizon ALFWorld tasks with Qwen3‑14B while preserving correct observations and demonstrating that recovery’s benefit is not solely due to new observations.
By Shuyao Xiao, Shengling Wang, Xuan Chen, Ke Chao, Ming Cui, Feifei Qian, Chaoyang Mei, Fanlin Meng, Ziming Yu, Junxi Yin
Legal Research Bench (LRB) is a new benchmark comprising 413 open-ended U.S. legal research questions, each paired with a gold answer, supporting authorities, and a binary grading rubric. The study evaluates thirteen advanced language‑model agents using web search, case‑law search, page parsing, and retrieval tools, scoring responses only when all required criteria are met and cited authorities verify. Results show that even the best model, Claude Opus 4.8, achieves full correctness on only 42.9% of questions, with performance varying by legal area and task complexity, and no clear link between more tool calls or inference cost and higher accuracy.
By Katrina Drozdov, Oliver Chen, Langston Nashold, Rayan Krishnan
The paper introduces a meta-multi-agent reinforcement learning (meta‑MARL) framework that enables rapid adaptation of interactive policies in multi‑agent systems. By modeling multi‑agent reinforcement learning problems as Markov games and defining a new concept called meta‑NE, the authors establish conditions linking meta‑NE to stationary points of a gradient‑play meta‑MARL algorithm. Experiments on autonomous‑driving tasks show that this approach adapts faster than pretrained MARL baselines, demonstrating its effectiveness.
By Huiwen Yan, Kyriakos G. Vamvoudakis, Mushuang Liu
ReLiveGym is a diagnostic environment that evaluates long‑lived language‑model agents over weeks of chronologically replayed real‑world streams such as news, market data, and social media. The tasks vary in time sensitivity, reasoning depth, and recurrence, and the study tests eight base language models to see how model choice and harness design—especially action timing—affect performance. Continuous learning from hindsight feedback is also examined to address failure modes in these long‑term tasks.
By Xisen Jin, Jingheng Li, Zhenglun Chen, Junyi Du, Xiang Ren
The paper introduces Generalized Implicit Temporal Abstraction (GITA), a method for goal-conditioned reinforcement learning that conditions a single value function on multiple temporal abstraction levels (k). By aggregating advantage-weighted supervision across various k values, GITA preserves both long-range signal and local resolution without committing to a single k. Experiments on OGBench show that GITA outperforms existing offline GCRL baselines, improving average success rates by 25 percentage points over HIQL and 7 percentage points over OTA.
By Pedro Robles Dutenhefner, Dikshant Shehmar, Wagner Meira Jr., Marlos C. Machado
ActiveSaddler introduces automated curriculum learning for harness optimization, treating the evolving training curriculum as a non‑stationary bandit problem. It identifies reusable failure patterns, estimates learning progress for each, and balances revisiting known weaknesses with exploring new scenarios, allowing the curriculum to co‑evolve with the harness. Experiments on GAIA2 and Terminal‑Bench 2.0 show consistent improvements in harness performance, with Pass@1 gains of 4.4 and 7.5 percentage points over fixed‑order baselines.
By Sungho Park, Wonjoong Kim, Jue Zhang, Wook-Shin Han, Pengfei Gao, Chanyoung Park, Yongqiang Yao, Rao Fu, Elsie Nallipogu, Qingwei Lin, Victor R\"uhle
The study evaluates 66 combinations of four configurable harnesses and five language models across three benchmark suites, revealing that model rankings and best harnesses vary significantly with the task. Notably, the openJiuwen harness consistently yields the highest scores for the Kimi model, while GPT performs best with the lean PI scaffold on Terminal‑Bench 4. The findings demonstrate that a model’s performance is highly dependent on the harness it is paired with, and that higher cost or a vendor’s own harness does not guarantee superior results.
By Yixuan Li, Yiyun Zhou, Yao Long Teng, Fuchao Yang, Yanchen Deng, Zhiyi Lyu, Xuyu Dong, Feng Chen, Bo An
The paper argues that as AI systems increasingly generate code, the bottleneck has shifted to supervising these systems, revealing a vocabulary gap between cybernetic coordination (actions aligning with the world) and epistemic coordination (understanding that can be verified). It critiques current oversight that merely approves outputs, proposing instead that every consequential choice by an agent must include a retrievable condition explaining why it was made, enabling third‑party verification. The authors illustrate this with three delegation episodes, introduce a two‑part reconstruction test, and propose the ORRCF convention to embed such conditions in all recorded decisions.
By J\'er\'emie Lumbroso
The paper introduces InFlowOp, a label‑free optimization framework that assigns costs to each decision in a multi‑agent workflow, balancing agent competence against execution time. It determines task granularity and agent assignment before execution and corrects faults during execution using the same cost metric. The authors also present Braid, a benchmark for multi‑agent coordination, and show that InFlowOp outperforms single‑agent baselines by up to 11.97% across various domains.
By Xuehang Guo, Haoyu Wang, Shengyu Chen, Zach Chen, Wei Cheng, Qingyun Wang, Haifeng Chen
The paper introduces Independent–Communicate–Revise (ICR), a framework that isolates communication effects in large language model multi‑agent systems by fixing initial reasoning and measuring how messages influence answer revision. ICR evaluates correction, preservation, and selectivity across four reasoning benchmarks, revealing that similar overall accuracy can mask divergent revision behaviors. The study shows that richer messages can both improve and harm outcomes, and that receiver policies can shift preservation and correction dynamics differently across tasks.
By Shixuan Li, Wei Yang, Peiyu Zhang, Anzhe Cheng, Heng Ping, Paul Bogdan
YouRA (Your Research Agent) is a persistent-state architecture that enables evidence‑traceable autonomous research agents to maintain research state, execution evidence, and failure history across long‑horizon pipelines. It combines a Verification State Architecture to track hypotheses and evidence, an Independent Controller to manage lifecycle and recovery, and Stateful Reflection to log failures and guide repair. On the MLR‑Bench ten‑task subset, YouRA outperforms MLR‑Agent and AI Scientist V2, and ablation studies confirm the importance of each component.
By Yoonkyu Woo, Woojin Lee, Jin-Xia Huang
The paper introduces a typed snapshot‑settlement contract for auditing concurrent actions in large language model agent environments. It evaluates three properties—order sensitivity, useful progress, and replay consistency—across five settlement policies, using 28,800 exhaustive permutation trials and 2,160 scripted multistep episodes. Results show that joint policies are spatially order‑invariant with fixed priorities, but conservative rejection only completes 31.25% of agents in a six‑agent doorway task compared to 90.28% for random tickets, while a full‑state journal audit successfully replays 156 checkpoints and rejects 1,332 constructed corruptions.
By Haotian Chen, Bowen Ye, Yuning Zhang, Jingkun Yu
The paper introduces Federated Agent Optimization (FAO), a framework for enabling large language model agents to improve collaboratively while keeping raw data, trajectories, and private knowledge local. FAO treats agent capabilities—such as memory, tools, rewards, skills, and structured knowledge—as a multi‑objective optimization space that balances utility, privacy leakage, and communication cost. It outlines methods for abstracting, protecting, aggregating, and adapting private experience into transferable capabilities, and highlights key challenges and future research directions for trustworthy federated agent systems.
By Qiang Yang, Zhiqiang Kou, Xueyi Zhang, Dong-Dong Wu, Hanlin Gu, Jing Guo, Yang Liu, Di Jiang, Qian Xu
The paper introduces an educator‑guided large language model (LLM) pedagogical agent that provides scaffolded feedback for conceptual database design. Embedded in an entity–relationship diagram editor, the system bases its feedback on the student’s artifact, assignment requirements, educator‑authored rubrics, and instructional resources, using a four‑stage workflow that progresses from concept checks to low‑detail feedback and localized clarification. In a deployment across three ERD environments with 383 feedback episodes, 71.1% of target‑level changes incorporated the hidden diagnostic target, and qualitative analysis highlighted the benefits and challenges of staged disclosure.
By Sara Riazi, Pedram Rooshenas
DeFA is a dependency-guided framework that attributes failures in large language model agents by constructing an event dependency graph and a failure propagation graph from protocol relations and semantic dependencies. It identifies violating events, traces their sources and effects, and determines the decisive error, responsible agent, and error category. The method supports long trajectories through segmentation and has shown superior accuracy on text, image, and video tasks, while its diagnostic feedback can improve agent performance on subsequent tasks.
By Bo Deng, Xinlei Zheng, Yi Wei, Kang Zhou, Chongyang Tao, Renzhao Liang, Xuanren Chen, Lifan Guo, Chi Zhang
The paper proposes a claim‑specific verification audit for modular agents that replaces aggregate task scores with evidence‑based evaluations. Each agent conclusion is recorded with supporting evidence and classified as supported, unsupported, unresolved, or not evaluated, along with the boundary of validity. The audit employs three tools—oracle policies, perfect component replacements, and verifier‑score tests—to trace value changes, locate lost value, and assess verifier effectiveness, demonstrated on a portfolio‑allocation agent in a synthetic market.
By Ali Atiah Alzahrani
The paper introduces a multi‑agent large language model framework that interprets personalized health checkup results by assigning distinct tasks to specialized agents and combining their outputs. Experiments on 120 Korean compound queries show that the multi‑agent approach improves overall quality scores and user preference compared to a single‑agent baseline, though it incurs higher latency and cost. The benefits are most pronounced for queries requiring personal‑record lookup.
By HyungJun Kim, Taehan Lee, Soojin Cheon
The paper evaluates five vision‑language models on autonomous driving tasks under various visual input conditions, finding that visual corruption affects accuracy and confidence differently across models and datasets. It then tests Visual Evidence Augmentation (VEA) as an inference‑time technique to enhance reliability, observing mixed improvements depending on the model and setting.
By Manasa Mariam Mammen, Priyanka Mary Mammen, Zafer Kayatas, Stefan Wagner