Agent-ValueBench is the first dedicated benchmark for agent values, showing they diverge from LLM values, form a homogeneous 'Value Tide' across models, and bend under harnesses and skill steering.
hub
Meta Context Engineer- ing via Agentic Skill Evolution, February 2026.https://arxiv.org/abs/2601.21557
15 Pith papers cite this work. Polarity classification is still indexing.
15
Pith papers citing it
hub tools
citation-role summary
background 1
baseline 1
method 1
citation-polarity summary
years
2026 15representative citing papers
Framework estimates context-dependent marginal utility of candidate skills via reward gaps in matched base vs. skill-augmented rollouts to filter skills and co-train policy as generator.
Evolutionary coding agents achieve most benchmark gains through a small subset of edit types and by cycling previously deleted code lines rather than developing new algorithmic structures.
Meta-Harness discovers improved harness code for LLMs via agentic search over prior execution traces, yielding 7.7-point gains on text classification with 4x fewer tokens and 4.7-point gains on math reasoning across held-out models.
Different patterns of environmental change (space vs time) require different LLM update behaviors; no single family of methods—prompts, distillation, RL, or compression—handles all regimes.
Probe-and-refine tuning refines AGENTS.md files using synthetic probes and improves coding agent resolve rate on SWE-bench Verified from 28.3% to 33.0% mainly by increasing coverage rather than per-patch precision.
Agent-guided prompt search with diagnostic code execution does not dominate simpler search on benchmarks, but in a production mental-health chatbot it chained eight noisy A/B tests into a claimed +29.4% next-day-retention gain.
HORMA builds a hierarchical memory structure from agent experiences and trains a lightweight RL navigator to retrieve minimal sufficient context, yielding better task performance with at most 22.17% of baseline token usage on ALFWorld, LoCoMo, and LongMemEval.
DemoEvolve bootstraps harness evolution with demonstrations to achieve more stable and effective edits than self-rollout search in sparse-feedback environments like Balatro.
SBD is a bilevel optimization framework that learns context-dependent safety weights for runtime task delegation in hierarchical multi-agent systems, with continuous authority transfer alpha and theoretical guarantees on safety monotonicity, policy convergence, and accountability propagation.
Across 2,923 GitHub repositories, context files - especially AGENTS.md and CLAUDE.md - dominate agentic-tool configuration, while advanced mechanisms like Skills and Subagents are rarely used and mostly static.
ConMem distills agent trajectories into structured memory cards organized in a relation-aware graph to enable training-free, relation-coordinated adaptation in LLM-based multi-agent systems.
Declarative planning in the harness accounts for the bulk of performance (+24.1pp win rate) while the LLM activates on only 4.3% of turns with bounded effect.
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
-
Agent-ValueBench: A Comprehensive Benchmark for Evaluating Agent Values
Agent-ValueBench is the first dedicated benchmark for agent values, showing they diverge from LLM values, form a homogeneous 'Value Tide' across models, and bend under harnesses and skill steering.
-
Co-Evolving Skill Generation and Policy Optimization
Framework estimates context-dependent marginal utility of candidate skills via reward gaps in matched base vs. skill-augmented rollouts to filter skills and co-train policy as generator.
-
What Do Evolutionary Coding Agents Evolve?
Evolutionary coding agents achieve most benchmark gains through a small subset of edit types and by cycling previously deleted code lines rather than developing new algorithmic structures.
-
Meta-Harness: End-to-End Optimization of Model Harnesses
Meta-Harness discovers improved harness code for LLMs via agentic search over prior execution traces, yielding 7.7-point gains on text classification with 4x fewer tokens and 4.7-point gains on math reasoning across held-out models.
-
When Does Continual Learning Require Learning
Different patterns of environmental change (space vs time) require different LLM update behaviors; no single family of methods—prompts, distillation, RL, or compression—handles all regimes.
-
Probe-and-Refine Tuning of Repository Guidance for Coding Agents
Probe-and-refine tuning refines AGENTS.md files using synthetic probes and improves coding agent resolve rate on SWE-bench Verified from 28.3% to 33.0% mainly by increasing coverage rather than per-patch precision.
-
SAGE: Stochastic Prompt Optimization via Agent-Guided Exploration
Agent-guided prompt search with diagnostic code execution does not dominate simpler search on benchmarks, but in a production mental-health chatbot it chained eight noisy A/B tests into a claimed +29.4% next-day-retention gain.
-
Organize then Retrieve: Hierarchical Memory Navigation for Efficient Agents
HORMA builds a hierarchical memory structure from agent experiences and trains a lightweight RL navigator to retrieve minimal sufficient context, yielding better task performance with at most 22.17% of baseline token usage on ALFWorld, LoCoMo, and LongMemEval.
-
DemoEvolve: Overcoming Sparse Feedback in Agentic Harness Evolution with Demonstrations
DemoEvolve bootstraps harness evolution with demonstrations to achieve more stable and effective edits than self-rollout search in sparse-feedback environments like Balatro.
-
Safe Bilevel Delegation (SBD): A Formal Framework for Runtime Delegation Safety in Multi-Agent Systems
SBD is a bilevel optimization framework that learns context-dependent safety weights for runtime task delegation in hierarchical multi-agent systems, with continuous authority transfer alpha and theoretical guarantees on safety monotonicity, policy convergence, and accountability propagation.
-
Harness Engineering for Agentic AI Coding Tools: An Exploratory Study
Across 2,923 GitHub repositories, context files - especially AGENTS.md and CLAUDE.md - dominate agentic-tool configuration, while advanced mechanisms like Skills and Subagents are rarely used and mostly static.
-
ConMem: Structured Memory-Guided Adaptation in Training-Free Multi-Agent Systems
ConMem distills agent trajectories into structured memory cards organized in a relation-aware graph to enable training-free, relation-coordinated adaptation in LLM-based multi-agent systems.
-
How Much Heavy Lifting Can an Agent Harness Do?: Measuring the LLM's Residual Role in a Planning Agent
Declarative planning in the harness accounts for the bulk of performance (+24.1pp win rate) while the LLM activates on only 4.3% of turns with bounded effect.
-
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.
- SkillMOO: Multi-Objective Optimization of Agent Skills for Software Engineering