mlsanityData to 14 Oct 2026

ATLAS-RTC: Closing the Loop on LLM Agent Output with Token-Level Runtime Control

arXiv 2603.27905 ↗First posted 29 Mar 2026cs.LGcited by 0

On the map: LLM agents and tool use → Software engineering coding agents

Abstract (arXiv)

We present ATLAS-RTC, a runtime control system for autoregressive language models that enforces structured output during decoding. ATLAS-RTC monitors generation at each step, detects drift from output contracts using lightweight signals, and applies targeted interventions such as biasing, masking, and rollback. Unlike post-hoc validation or static constrained decoding, it operates in a closed loop, enabling correction before errors materialize. Across structured generation and tool-calling tasks, ATLAS-RTC improves first-attempt success rates by 20 to 37.8 percentage points, with up to 88% latency reduction in failure-dominated settings. Results show that many failures arise from decoding artifacts rather than task misunderstanding, motivating runtime control as a distinct layer in LLM systems.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use. Builds on 12 of its 17 references that are in the map. First posted on arXiv 29 Mar 2026.

Builds on

12 of 17 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025VIGIL: A Reflective Runtime for Self-Healing Agents8 Dec

Other references · 11

  1. 2021PICARD: Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Modelscited by 411
  2. 2025AgentSpec: Customizable Runtime Enforcement for Safe and Reliable LLM Agentscited by 223
  3. 2023Grammar-Constrained Decoding for Structured NLP Tasks without Finetuningcited by 197
  4. 2024Guiding LLMs The Right Way: Fast, Non-Invasive Constrained Generationcited by 101
  5. 2024Grammar-Aligned Decodingcited by 54
  6. 2025CRANE: Reasoning with constrained LLM generationcited by 38
  7. 2026Towards Verifiably Safe Tool Use for LLM Agentscited by 36
  8. 2025ProbGuard: Proactive Runtime Monitoring for LLM Agent Safety via Probabilistic Predictioncited by 28
  9. 2025RvLLM: LLM Runtime Verification with Domain Knowledgecited by 17
  10. 2026The Hidden Cost of Structured Generation in LLMs: Draft-Conditioned Constrained Decodingcited by 7
  11. 2025Adaptive Focus Memory for Language Modelscited by 6

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025DecoRTL: A Run-time Decoding Framework for RTL Code Generation with LLMs
  2. 2026Structured Feedback Improves Repair in an LLM Agent Loop
  3. 2026Chronicle: Cut-Point Replay for Regression Testing of LLM Agents
  4. 2025Collab: Controlled Decoding using Mixture of Agents for LLM Alignment
  5. 2026CASS-RTL: Correctness-Aware Subspace Steering for RTL Generation with LLMs
  6. 2026AsymSpec: Context-Asymmetric Speculative Decoding for Agentic LLMs
  7. 2024GoEX: Perspectives and Designs Towards a Runtime for Autonomous LLM Applications
  8. 2026AgentPerfBench: A Benchmarking and Evaluation Suite for Inference Performance of Agentic LLMs