arXiv:2607. 00297v1 Announce Type: new Abstract: When LLM agents use evaluator feedback to adapt their behavior in closed loops, evaluator biases propagate through the agent's strategy distribution -- a phenomenon known as evaluator preference coupling.
arXiv:2606. 31371v1 Announce Type: cross Abstract: When large language model (LLM) agents adapt their behavior through evaluator feedback, systematic evaluator biases propagate into the agent's learned strategy distribution - a phenomenon termed evaluator preference coupling.
arXiv:2609.26693v1 Announce Type: new Abstract: A coding agent must emit a valid tool call--a parseable invocation of a tool in the provided schema--before the harness can execute its chosen action....
arXiv:2606. 17546v1 Announce Type: new Abstract: Self-evolving LLM-based agents improve mainly by changing their agent harness: the structured execution layer around a base model, including prompts, memory, tools, middleware, runtime state, and the model-tool interaction loop.
arXiv:2606. 29719v1 Announce Type: new Abstract: Measurements of proprietary LLM evaluators can become invalid within weeks -- we document one case and provide the diagnostic framework to detect it.
arXiv:2605. 27898v2 Announce Type: replace Abstract: As LLMs are increasingly deployed as agents, reliable assessment of their agentic capabilities has become essential.
UniACE is a unified framework that standardizes the evaluation of large language model (LLM) agents by representing each benchmark as an instruction–tool–environment triplet and running models through a shared, task‑agnostic harness in isolated runtimes. It preserves native success criteria, offers an offline mode for dynamic‑resource tasks, and standardizes efficiency metrics, execution records, and failure attribution. Applying UniACE to 7 benchmarks across 24 domains and 15 models revealed significant score shifts, ranking reversals, and sensitivity to evidence representation, highlighting the impact of evaluation configuration on reported agent performance.
arXiv:2607. 16345v1 Announce Type: cross Abstract: Modern agentic systems increasingly rely on skills: installable packages of natural language and code that teach an LLM agent to perform a domain task.
arXiv:2604. 16706v2 Announce Type: replace Abstract: Automated evaluation of tool-using large language model (LLM) agents is widely assumed to be reliable, yet this assumption is rarely validated against human annotation.
arXiv:2604. 00392v2 Announce Type: replace-cross Abstract: Agents that synthesize their own tools ship a second artifact alongside each answer: a software library that future tasks reuse, compose, and depend on.
arXiv:2606. 05670v1 Announce Type: new Abstract: Does adding more agents help an LLM workflow once compared systems share the same benchmark loader, tool access, answer contract, usage accounting, and trajectory logging?
The paper introduces a framework for evaluating how large language model agents revise their success criteria after failures, defining five non‑compensatory conditions that must be met for a criterion revision to be considered valid. Using the CMB‑0.1 protocol, the authors test twelve cross‑domain scenarios across four system configurations, finding that no model trial satisfies all five conditions and highlighting specific failure modes such as zero‑state reconstruction and inadequate intervention sensitivity. They propose a more stringent trace‑anchored CMB‑0.4 protocol to better isolate and measure criterion revision in future studies.
arXiv:2608. 00881v1 Announce Type: new Abstract: Large language model agents increasingly act through stateful tools, yet model generation and environment execution remain serialized at every step.