Self-GC governs agent context as indexed objects with planner-proposed actions, achieving 84.85% no-impact on future continuations on a hard set versus 54-70% for baselines.
Improving the efficiency of llm agent systems through trajectory reduction
11 Pith papers cite this work. Polarity classification is still indexing.
11
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 11roles
background 1polarities
background 1representative citing papers
AGORA is an inference-free step-level compressor for LLM agent prompts that retains at least 75% of uncompressed performance in most tested settings where token-level methods collapse due to action-grammar destruction.
A local Llama 3.2 pre-flight rewriter that translates and compacts non-English coding prompts cuts prompt tokens by 34–47% on a new 200-task benchmark while preserving accuracy across three commercial backends.
SAM is a standalone memory framework for long-horizon LLM agents that creates state-adaptive cues from interactions, preserves raw trajectories for intent-driven recall, and optimizes the module via expert supervision and RL, outperforming baselines on BrowseComp and related benchmarks.
SieveFL combines vector retrieval and JaCoCo runtime pruning to cut LLM token use by 49% while achieving 41.8% Top-1 accuracy on 395 Defects4J bugs, outperforming AgentFL.
An LLM agent with decompiler, COM-inspection, and debugger tools finds race-condition vulnerabilities in Windows COM binaries and generates verified proof-of-concept crashes.
QuantClaw dynamically routes precision in agent workflows to cut cost by up to 21.4% and latency by 15.7% while keeping or improving task performance.
SWE-MeM introduces adaptive memory management for coding agents via synthesized trajectories and Memory-aware GRPO, reporting 43.4% and 60.2% resolve rates on SWE-Bench Verified for 4B and 30B models while beating baselines on performance and token use.
HarnessFix diagnoses harness flaws from agent traces via HTIR, maps them to repair operators, and improves benchmark performance by 6.3-18.4% over baselines.
Derives unique closed-form decentralized policy minimizing worst-agent online regret that asymptotically converges to centralized Nash-optimal policy in mean-field limit, with added online mixture weighting.
Code minification reduces average input token usage by 42% in state-in-context agents with a 12 percentage point drop in resolution rate on SWE-bench Verified.
citing papers explorer
-
Self-GC: Self-Governing Context for Long-Horizon LLM Agents
Self-GC governs agent context as indexed objects with planner-proposed actions, achieving 84.85% no-impact on future continuations on a hard set versus 54-70% for baselines.
-
AGORA: Adapter-Grounded Observation-Action Retention for Inference-Free Prompt Compression in LLM Agents
AGORA is an inference-free step-level compressor for LLM agent prompts that retains at least 75% of uncompressed performance in most tested settings where token-level methods collapse due to action-grammar destruction.
-
Cross-Lingual Token Arbitrage: Optimizing Code Agent Context Windows via Local LLM Preprocessing
A local Llama 3.2 pre-flight rewriter that translates and compacts non-English coding prompts cuts prompt tokens by 34–47% on a new 200-task benchmark while preserving accuracy across three commercial backends.
-
SAM: State-Adaptive Memory for Long-Horizon Reasoning Agent
SAM is a standalone memory framework for long-horizon LLM agents that creates state-adaptive cues from interactions, preserves raw trajectories for intent-driven recall, and optimizes the module via expert supervision and RL, outperforming baselines on BrowseComp and related benchmarks.
-
SieveFL: Hierarchical Runtime-Aware Pruning for Scalable LLM-Based Fault Localization
SieveFL combines vector retrieval and JaCoCo runtime pruning to cut LLM token use by 49% while achieving 41.8% Top-1 accuracy on 395 Defects4J bugs, outperforming AgentFL.
-
Agentic Vulnerability Reasoning on COTS Binaries
An LLM agent with decompiler, COM-inspection, and debugger tools finds race-condition vulnerabilities in Windows COM binaries and generates verified proof-of-concept crashes.
-
QuantClaw: Precision Where It Matters for OpenClaw
QuantClaw dynamically routes precision in agent workflows to cut cost by up to 21.4% and latency by 15.7% while keeping or improving task performance.
-
SWE-MeM: Learning Adaptive Memory Management for Long-Horizon Coding Agents
SWE-MeM introduces adaptive memory management for coding agents via synthesized trajectories and Memory-aware GRPO, reporting 43.4% and 60.2% resolve rates on SWE-Bench Verified for 4B and 30B models while beating baselines on performance and token use.
-
From Failed Trajectories to Reliable LLM Agents: Diagnosing and Repairing Harness Flaws
HarnessFix diagnoses harness flaws from agent traces via HTIR, maps them to repair operators, and improves benchmark performance by 6.3-18.4% over baselines.
-
MEMOA: Massive Mixtures of Online Agents via Mean-Field Decentralized Nash Equilibria
Derives unique closed-form decentralized policy minimizing worst-agent online regret that asymptotically converges to centralized Nash-optimal policy in mean-field limit, with added online mixture weighting.
-
Reducing Token Usage of State-in-Context Agents using Minification
Code minification reduces average input token usage by 42% in state-in-context agents with a 12 percentage point drop in resolution rate on SWE-bench Verified.