Related papers: Ratchet: A Minimal Hygiene Recipe for Self-Evolvin…

Conducting literature reviews for scientific papers is essential for understanding research, its limitations, and building on existing work. It is a tedious task which makes an automatic literature review generator appealing. Unfortunately,…

Code production is now a commodity; the bottleneck is knowing what to build and proving it works. We present the Kitchen Loop, a framework for autonomous, self-evolving software built on a unified trust model: (1) a specification surface…

Small LLMs often struggle to match the agentic capabilities of large, costly models. While reinforcement learning can help, progress has been limited by two structural bottlenecks: existing open-source agentic training data are narrow in…

As LLM agents are increasingly built around reusable skills, a central challenge is no longer only whether agents can use provided skills, but whether they can generate correct, reusable, and executable skills from repositories and…

We present RELISH (REgression with a Latent Iterative State Head), a novel, lightweight architecture designed for text regression with large language models. Rather than decoding numeric targets as text or aggregating multiple generated…

With LLMs shifting their role from statistical modeling of language to serving as general-purpose AI agents, how should LLM evaluations change? Arguably, a key ability of an AI agent is to flexibly combine, as needed, the basic skills it…

Online health resources and large language models (LLMs) are increasingly used as a first point of contact for medical decision-making, yet their reliability in healthcare remains limited by low accuracy, lack of transparency, and…

As LLM-based agents increasingly rely on external tools, it is important to evaluate their ability to sustain tool-grounded reasoning beyond familiar workflows and short-range interactions. We introduce AgentEscapeBench, an…

Reward Models (RMs) are crucial to aligning large language models (LLMs), but the degree to which an RM specialized to one task (e.g. writing) generalizes to new tasks (e.g. math) is often not known a priori, often making using only one…

Agent Skills are structured packages of procedural knowledge that augment LLM agents at inference time. Despite rapid adoption, there is no standard way to measure whether they actually help. We present SkillsBench, a benchmark of 86 tasks…

Large Language Model (LLM)-based agents have demonstrated remarkable capabilities in complex reasoning and multi-turn interactions but struggle to continuously improve and adapt when deployed in new environments. One promising approach is…

Large language models (LLMs) exhibit remarkable performance across various NLP tasks. However, they often generate incorrect or hallucinated information, which hinders their practical applicability in real-world scenarios. Human feedback…

Current Large Language Model (LLM) agents show strong performance in tool use, but lack the crucial capability to systematically learn from their own experiences. While existing frameworks mainly focus on mitigating external knowledge gaps,…

Recent breakthroughs in large language models (LLMs) on reasoning tasks rely heavily on massive, high-quality datasets-typically human-annotated and thus difficult to scale. While data synthesis or distillation offers a promising…

For sparse, structured reinforcement-learning tasks with semantic reward-function interfaces, LLM-generated reward shaping is better framed as debugging than one-shot generation. We study PPO-trained agents using MiniGrid as core evaluation…

Single-agent LLMs hit hard limits--finite context, role overload, and brittle domain transfer. Conventional multi-agent fixes soften those edges yet expose fresh pains: ill-posed decompositions, fuzzy contracts, and verification overhead…

Supervised fine-tuning (SFT) of large language models (LLMs) for specialized tasks requires high-quality datasets, but manual curation is prohibitively expensive. Synthetic data generation offers scalability, but its effectiveness relies on…

Agent skills, which are reusable, domain-specific knowledge artifacts, have become a popular mechanism for extending LLM-based agents, yet formally benchmarking skill usage performance remains scarce. Existing skill benchmarking efforts…

Despite the rapid advancements in LLM agents, they still face the challenge of generating meaningful reflections due to inadequate error analysis and a reliance on rare successful trajectories, especially in complex tasks. In this work, we…

Skill-distillation pipelines learn reusable rules from LLM agent trajectories, but they lack a key signal: how much each step costs. Without per-step cost, a pipeline cannot distinguish adding a missing step to fix a bug from removing an…