The paper introduces MAS (Model‑Agnostic Selection with agent‑X importance), a new method for controlling the false discovery rate (FDR) in variable selection that does not rely on specific model assumptions. MAS uses sample splitting, a novel agent‑X importance measure, a resplit‑per‑covariate strategy, and data aggregation to achieve asymptotic FDR control while improving computational efficiency and power. The authors provide theoretical guarantees and demonstrate competitive performance through extensive simulations and real‑data analyses.
By Mengtao Wen, Haojie Ren, Changliang Zou
CausalEvo is a method that enables large language model agents in health to evolve their textual skills—such as calculation rules and measurement units—by evaluating the causal impact of skill changes while keeping model weights, feedback cases, and rule count constant. The approach freezes the LLM to propose candidate rules, pools them with existing skills, and uses paired skill replacements scored on shared feedback to rank and retain the most effective rules. In experiments on six MedCalc-Bench-Verified tasks, CausalEvo achieved 60.00% accuracy, outperforming SkillOpt (55.83%) and a no‑skill baseline (51.67%), while also reducing training token usage by 55.6%.
whyItMatters":"CausalEvo demonstrates that controlled, causal evaluation of skill updates can improve health‑focused LLM performance and efficiency, directly benefiting the reliability and resource usage of medical language agents."
By Yuchen Zhang
MemComposer is a memory composer for large language model agents that selects, orders, and limits memories based on the current query and retrieved candidates. It is trained with Experience Policy Optimization (ExPO), a GRPO-based framework that evaluates assembled memory contexts through full agent interactions and optimizes the assembly policy using task‑success feedback. In experiments on three τ²‑bench domains, MemComposer outperforms the strongest baseline for each executor while reducing memory‑token usage by 16.9% and 4.9%, and its learned policy transfers to unseen domains, executors, and memory banks.
By Yixuan Xie, Xueqiang Xu, Xinyi Fan, Xinxi Lyu, Tao Feng, Xing Fan, Chenlei Guo, Jie Hao, Shaowen Wang, Jiashuo Sun, Jiawei Han
Terminal‑RSI is a new framework for terminal agents that alternates Recursive Self‑Rewrite (RSR) of verified trajectories with Recursive Policy‑Refine (RPR) that inherits a critic. It introduces TB‑GAE for token‑specific baselines, Family Dense Reward (FDR) and Success‑Prioritized Reward (SPR) to calibrate progress, and uses a warm‑up critic, recursive handoff, and stage‑specific learning rates to coordinate the two refinement phases. The five‑stage pipeline applied to Qwen3.8‑27B produces Terminal‑RSI‑27B, achieving 19.84 % on Terminal‑Bench 4.0, 86.52 % on Terminal‑Bench 2.1, 16.95 % on Terminal‑Bench 3.0, and 79.00 % on Terminal‑Bench Hard, while improving process reward and strict completion on LHTB.
By Junyao Yang, Zongxia Li, Yucheng Shi, Zhongzhi Li, Ruhan Wang, Ruiwen Zhou, Haitao Mi, Shuicheng Yan, Leowei Liang
Memorilla is a memory system that encodes an entire document collection into reusable embeddings and retrieves query‑specific information as a fixed set of memory tokens for a frozen large language model. Using only 16 memory tokens, it outperforms baselines such as Mem0 and MemoRAG on most benchmarks, achieving higher accuracy on FactKG while drastically reducing token usage. In agentic settings, Memorilla helps prevent wandering and boosts success rates in a TextWorld environment, demonstrating its utility for continual learning and long‑context interactions.
By Adibvafa Fallahpour, Parsa Idehpour, Vignesh Kothapalli, Nikita Mounier, Shirley Wu, Shayan Pardis, Jure Leskovec
The paper introduces a benchmark of 400 synthetic healthcare workflow episodes to evaluate large language model (LLM) agents’ reliability in escalating care when evidence is missing or conflicting. It finds that while clean decision accuracy is high (0.97), models miss a large proportion of necessary escalations (32–97%) under uncertainty, and even the largest 9.7B model’s performance drops to 0.78 in conflict scenarios. A workflow-aware controller improves utility and reduces missed escalations, but its effectiveness is limited by lexical generalization failures.
By Zihao Zheng (Washington University in St. Louis), Junyi Yao (Washington University in St. Louis), Baichuan Li (Southern Methodist University), Jiayu Long (Washington University in St. Louis)
Research: sparqlite: SPARQL 1.1 inside SQLite, written in Rust
This research task is really an experiment in running Claude Code for web in "Ultracode" mode, as described in Orchestrate subage...
Stack Overflow for Agents was launched in just a few weeks by two engineers using Codex, and its ChatGPT plugin contributed to up to tenfold growth in agent creation.
See how Localiza uses ChatGPT Enterprise, the OpenAI API, Codex, and 60+ specialized agents to reorganize work and build AI-native mobility experiences.
Almost 50 years after starting in a compounding pharmacy in Paraná, the group operates a proprietary agent platform that autonomously resolves up to 80% of resellers’ operational requests—enabling the...
As assistants move from reactive answering toward proactive intervention, memory changes role: it no longer only supplies the answer to a query, it determines whether and when to speak at all. Prior w...
MLExpert is a 35B Mixture-of-Experts agentic model designed to autonomously conduct long‑horizon machine learning engineering (MLE) tasks. It uses a post‑training approach that combines hierarchical supervised fine‑tuning with Windowed Interaction RL to learn from intermediate outcomes in complex experimentation workflows. On the MLE‑Bench Lite benchmark, MLExpert raises the average medal rate from 37.9% to 72.7%, matching the performance of larger frontier models and demonstrating transferable optimization capabilities to other benchmarks such as NatureBench Lite.
The paper investigates how to improve environment‑agnostic coding agents by refining prompts and leveraging scaffolding. By modifying a single instruction to encourage reporting difficulties and allowing automatic continuation, the authors achieve higher mean scores on ARC‑AGI‑3 and AGI Maze, though performance on Terminal‑Bench remains mixed. A scripted controller that handles reset decisions recovers part of the specialized agent’s advantage, while additional tutorial experience can actually reduce performance, underscoring the importance of effectively using available information and tools.
Intern-Decision introduces 0.8B, 2B, and 4B multimodal decision models that output structured choices and probabilities directly from text and optional images, avoiding the latency of free‑form responses. The models use an autoregressive masked language modeling objective, placing decision placeholders in a JSON skeleton and evaluating all fields in a single causal pass. The 4B variant attains 90.02% unweighted accuracy across seven benchmarks, with sub‑45 ms latency on an RTX 4090, and demonstrates improved calibration and handoff performance in pilot studies.
Behavioral consistency across repeated executions provides an informative signal for assessing the reliability of large language model (LLM) agents. However, existing approaches typically summarize co...
As multimodal AI systems are becoming our everyday thought partners, the shift towards studying how users transfer control to these systems becomes especially urgent. In this paper, we report an overv...
Despite the remarkable generalization capabilities of Large Language Models, generating clinically accurate and structured radiology reports from medical images remains a significant challenge. Curren...
Aerial vision-language navigation (VLN) agents built on large vision-language models are expensive to query. Executing a complete predicted action chunk before replanning amortizes this cost, reducing...
Large language model (LLM) agents now fix software, operate computers and serve customers, but their harnesses, the programs that run them, call a large LLM at every step, which makes long tasks slow...
Centralized orchestration provides comprehensive multi-agent coordination but creates bottlenecks in scalability, adaptability, and collective performance. Decentralized approaches alleviate these bot...