BenchTrace supplies a fixed snapshot-reflection dataset and two targeted evaluations plus the failure avoidance rate metric to measure reflection quality and controlled self-evolution in LLM agents.
arXiv preprint arXiv:2603.15255 , year =
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
years
2026 6roles
background 2polarities
background 2representative citing papers
MAGE uses a four-subgraph co-evolutionary knowledge graph plus dual bandits to externalize and retrieve experience for stable self-evolution of frozen language-model agents, showing gains on nine diverse benchmarks.
RLVR training teaches a 30B LLM to strategically explore a multi-buyer market and extract 70% of available surplus, outperforming frontier models up to 1T parameters in concurrent negotiation.
EVE-Agent adds an evidence verifier to the proposer-solver loop that rewards spans by marginal accuracy gain, producing self-generated but inspectable training examples for search agents.
MetaResearcher is a proposed multi-component framework for scaling deep research agent training via adversarial virtual worlds, discovery tasks, meta-rewards, and multi-agent collaboration.
Survey framing LLM agents as model-plus-harness systems, decomposing harness responsibilities, mapping them to tasks, and highlighting open challenges in evaluation, safety, and co-evolution.
citing papers explorer
-
BenchTrace: A Benchmark for Testing Reflection Ability and Controlled Evolution in LLM Agents
BenchTrace supplies a fixed snapshot-reflection dataset and two targeted evaluations plus the failure avoidance rate metric to measure reflection quality and controlled self-evolution in LLM agents.
-
MAGE: Multi-Agent Self-Evolution with Co-Evolutionary Knowledge Graphs
MAGE uses a four-subgraph co-evolutionary knowledge graph plus dual bandits to externalize and retrieve experience for stable self-evolution of frozen language-model agents, showing gains on nine diverse benchmarks.
-
Strategic Bargaining in Multi-Buyer Markets: Reinforcement Learning from Verifiable Rewards for LLM Negotiations
RLVR training teaches a 30B LLM to strategically explore a multi-buyer market and extract 70% of available surplus, outperforming frontier models up to 1T parameters in concurrent negotiation.
-
EVE-Agent: Evidence-Verifiable Self-Evolving Agents
EVE-Agent adds an evidence verifier to the proposer-solver loop that rewards spans by marginal accuracy gain, producing self-generated but inspectable training examples for search agents.
-
MetaResearcher: Scaling Deep Research via Self-Reflective Reinforcement Learning in Adversarial Virtual Environments
MetaResearcher is a proposed multi-component framework for scaling deep research agent training via adversarial virtual worlds, discovery tasks, meta-rewards, and multi-agent collaboration.
-
From Question Answering to Task Completion: A Survey on Agent System and Harness Design
Survey framing LLM agents as model-plus-harness systems, decomposing harness responsibilities, mapping them to tasks, and highlighting open challenges in evaluation, safety, and co-evolution.