Control Under Compression: Reliability Frontiers for Tool-Using Agents

• URL: http://arxiv.org/abs/2608.01056v1
• Date: Sun, 02 Aug 2026 07:43:44 GMT
• Title: Control Under Compression: Reliability Frontiers for Tool-Using Agents
• Abstract summary: We introduce CompressAgent, an environment-verified benchmark for ACC compression.<n>At 75% retained context, generic rewriting and section-based compression achieve 92.7% and 92.4% success.<n>At retained-context budgets from 25% to 10%, executable protocols become fragile.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Tool-using language-model agents are governed not only by task prompts but also by persistent system-side instructions that specify tools, arguments, policies, execution protocols, and recovery. Compressing these agent control contexts (ACCs) can reduce input cost and context use, yet existing prompt-compression evaluations do not reveal whether the resulting control remains operationally reliable. We introduce CompressAgent, an environment-verified benchmark for ACC compression across nine independently constructed ACCs, three task families, three fixed Qwen API model identifiers, six retained-context budgets, and 15,525 runs. We uncover a nonlinear, method-dependent reliability frontier. At 75% retained context, generic rewriting and section-based compression achieve 92.7% and 92.4% success, close to the 93.8% full-context baseline. Between 50% and 35%, methods diverge sharply; at 35%, section-based, obligation-aware, and generic rewriting achieve 47.0%, 39.0%, and 19.9%. At retained-context budgets from 25% to 10%, executable protocols become fragile. Reliability also varies substantially across ACCs, making universal compressor rankings inappropriate and motivating per-context qualification. Failure analysis shows that compression primarily surfaces as tool-execution and action-parsing errors. These findings recast ACC compression from token reduction into a runtime-reliability problem that must be evaluated through executable outcomes.

Related papers

• MemToC: Benchmarking Memory-Tool Conflict Resolution in Large Language Models [49.47300047353726]
  We introduce MemToC, a benchmark for post-tool-return arbitration with executable tools.<n>MemToC comprises 6,504 evaluation episodes constructed from 542 quality-controlled factual questions.<n>Across five open-weight 7-9B models, tool returns strongly dominate elicited closed-book answers.
  arXiv  Detail & Related papers  (2026-08-26T18:22:03Z)
• Task-Conditioned Least-Privilege Learning for Executable Terminal and MCP Agents [45.88028371034407]
  Tool-using large language-model agents can complete a task while exercising authority that the user did not grant or the task does not need.<n>We propose a framework where each action is audited before execution and again from observed effects along six dimensions of risk.<n>We find that after training using this framework on Qwen3.5-4B over 1,500 tasks, the selected seed reaches 98.48% safe success across 2,896 evaluation episodes.
  arXiv  Detail & Related papers  (2026-08-18T22:07:23Z)
• KVDiagnosis: A Diagnostic Benchmark for KV-Cache Compression in Long-Context Language Models [5.349156411852564]
  KV-cache compression reduces long-context memory, but aggregate task scores reveal neither which correct executions fail nor why.<n>We present KVDiagnosis, a diagnostic dataset and benchmark with three contributions.
  arXiv  Detail & Related papers  (2026-08-10T10:38:05Z)
• Claim Plane: Reliability Gains and the Limits of Selective Concurrency for Parallel Coding Agents: A 30-Pair, Three-Seed Confirmatory Study of Deterministic Pre-Write Admission [1.4721615285883427]
  Claim Plane addresses this failure mode as deterministic pre-write admission over versioned change intents.<n>Static Claim Plane raised pair pass from 23.3% under unconstrained parallel execution to 50.0%.<n> Dynamic admission was more selective, serializing 66.7% of conflict cases and 13.3% of clean cases, but 46 of 90 executions failed closed on undeclared scope.
  arXiv  Detail & Related papers  (2026-08-02T02:56:49Z)
• EduPluginBench: Executable Assurance for AI-Generated Educational Plugins [0.0]
  We introduce negative provenance, an executable benchmark and staged admission method for generated plugins in software ecosystems.<n>A frozen transfer study of 600 unmodified generations from two current coding models found that 300/600 parsed, but none passed P0 or achieved P0-P4 conformance.<n>An earlier 540-generation diagnostic found that post-hoc bounded repair yielded 112 P0 passes, all nonconforming, with recall increasing from 13.4% to 100%.
  arXiv  Detail & Related papers  (2026-08-01T16:22:35Z)
• Token Reduction Is Not Cost Reduction [0.0]
  We show that prompt-cache traffic dominated cost composition, accounting for about 87% of reconstructed four-component cost.<n>Local payload reduction was not a reliable predictor of end-to-end billed cost.<n>We propose evaluating context-reduction systems by success-adjusted billed cost rather than token reduction alone.
  arXiv  Detail & Related papers  (2026-07-13T21:10:22Z)
• Understanding and Evaluating Claw-like Agent Security Through a Computer-Systems Lens [65.53494487819053]
  Claw-like AI agents (e.g., OpenClaw) are always-on processes with persistent access to credentials, files, tools, and external services.<n>We adopt a computer-system analogy: treating a Claw-like agent as an agentic computer system whose gateway runtime plays an OS-like mediation role.<n>We develop SafeClawArena, a benchmark of 406 adversarial tasks across four attack surfaces.
  arXiv  Detail & Related papers  (2026-06-29T18:00:45Z)
• ACCORD: Action-Conditioned Contextual Grounding for Language Agents [69.73525608707764]
  ACCORD (Action-Conditioned Contextual Grounding) is an agent framework for adaptive grounding.<n>We show that current agents often fail to do so. They act from assumed rather than observed specifics, overlook information they could have gathered, and fail to incorporate evidence that has already been returned.
  arXiv  Detail & Related papers  (2026-06-15T09:05:55Z)
• Policy-Conditioned Counterfactual Credit for Verifiable Reinforcement Learning of Long-Horizon Language Agents [0.0]
  Reinforcement learning with verifiable rewards improves reasoning and tool use, yet long-horizon language agents still learn unsupported evidence chains and belief drift.<n>We propose a constrained policy-gradient algorithm with dense verifiable rewards, intervention-validity gating, and a policy-conditioned counterfactual contribution (PCCC) estimator.
  arXiv  Detail & Related papers  (2026-06-03T16:19:36Z)
• RAIL Guard: Closing the Evaluation-to-Remediation Gap in Responsible AI for LLM Agents [0.0]
  RAIL Guard is a closed-loop responsible AI pipeline.<n>It iteratively remediates failing outputs through an evaluate-rewrite-reevaluate loop.<n>System is available as open-source SDKs.
  arXiv  Detail & Related papers  (2026-05-28T21:23:49Z)
• Converted, Not Equivalent: Benchmarking Codebase Conversion via Observational Equivalence [56.25095230687242]
  Coding agents often over-trust their own local validation routines and declare success on artifacts that satisfy surface checks.<n>This problem is especially acute in conversion, where prior evaluation is outcome-driven.<n>Across blind conversion attempts, the best system reaches 26.7--28.9% overall pass rate despite Spec pass rates up to 91.1%.<n>This suggests that failures stem more from contract-misaligned self-validation than from limited budget or backbone strength.
  arXiv  Detail & Related papers  (2026-05-27T19:57:15Z)
• Credential Leakage in LLM Agent Skills: A Large-Scale Empirical Study [51.717224133855886]
  Third-party skills extend LLM agents with powerful capabilities but often handle sensitive credentials in privileged environments.<n>We analyze 17,022 skills (sampled from 170,226 on SkillsMP) using static analysis, sandbox testing, and manual inspection.<n>We identify 520 vulnerable skills with 1,708 issues and derive a taxonomy of 10 leakage patterns (4 accidental and 6 adversarial)
  arXiv  Detail & Related papers  (2026-04-03T14:50:16Z)
• From Black Box to Glass Box: Cross-Model ASR Disagreement to Prioto Review in Ambient AI Scribe Documentation [43.148402136307716]
  Cross-model disagreement among heterogeneous ASR systems can act as a reference-free uncertainty signal.<n>We transcribed 50 publicly available medical audio clips with eight ASR systems spanning commercial APIs and open-source engines.<n>Low-agreement regions were enriched for content disagreements, with the content fraction increasing from 53.9% to 73.9% across quintiles of high-risk mass.
  arXiv  Detail & Related papers  (2026-03-02T13:02:13Z)
• Trace: Securing Smart Contract Repository Against Access Control Vulnerability [58.02691083789239]
  GitHub hosts numerous smart contract repositories containing source code, documentation, and configuration files.<n>Third-party developers often reference, reuse, or fork code from these repositories during custom development.<n>Existing tools for detecting smart contract vulnerabilities are limited in their ability to handle complex repositories.
  arXiv  Detail & Related papers  (2025-10-22T05:18:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.