Breaking Contextual Inertia: Reinforcement Learning with Single-Turn Anchors for Stable Multi-Turn Interaction

• URL: http://arxiv.org/abs/2603.04783v1
• Date: Thu, 05 Mar 2026 04:04:59 GMT
• Title: Breaking Contextual Inertia: Reinforcement Learning with Single-Turn Anchors for Stable Multi-Turn Interaction
• Authors: Xingwu Chen, Zhanqiu Zhang, Yiwen Guo, Difan Zou,
• Abstract summary: We introduce textbfReinforcement textbfLearning with textbfTurn textbfRLSTA, a generalizable training approach designed to stabilize multi-turn interaction.<n>Experiments show that RLSTA significantly outperforms standard fine-tuning and abstention-based methods.
• Score: 49.03500737694832
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: While LLMs demonstrate strong reasoning capabilities when provided with full information in a single turn, they exhibit substantial vulnerability in multi-turn interactions. Specifically, when information is revealed incrementally or requires updates, models frequently fail to integrate new constraints, leading to a collapse in performance compared to their single-turn baselines. We term the root cause as \emph{Contextual Inertia}: a phenomenon where models rigidly adhere to previous reasoning traces. Even when users explicitly provide corrections or new data in later turns, the model ignores them, preferring to maintain consistency with its previous (incorrect) reasoning path. To address this, we introduce \textbf{R}einforcement \textbf{L}earning with \textbf{S}ingle-\textbf{T}urn \textbf{A}nchors (\textbf{RLSTA}), a generalizable training approach designed to stabilize multi-turn interaction across diverse scenarios and domains. RLSTA leverages the model's superior single-turn capabilities as stable internal anchors to provide reward signals. By aligning multi-turn responses with these anchors, RLSTA empowers models to break contextual inertia and self-calibrate their reasoning based on the latest information. Experiments show that RLSTA significantly outperforms standard fine-tuning and abstention-based methods. Notably, our method exhibits strong cross-domain generalization (e.g., math to code) and proves effective even without external verifiers, highlighting its potential for general-domain applications.

Related papers

• ALIVE: Awakening LLM Reasoning via Adversarial Learning and Instructive Verbal Evaluation [4.265094703231012]
  We introduce textbfALIVE (emphAdrial Learning with Instructive Verbal Evaluation), a hands-free alignment framework.<n>By coupling adversarial learning with instructive verbal feedback, ALIVE enables models to internalize evaluative criteria directly from raw corpora.<n>With identical data and compute, ALIVE achieves markedly improved cross-domain generalization, and higher self-correction rates.
  arXiv  Detail & Related papers  (2026-02-05T09:20:23Z)
• Guided Verifier: Collaborative Multimodal Reasoning via Dynamic Process Supervision [11.159231524113764]
  Reinforcement Learning (RL) has emerged as a pivotal mechanism for enhancing the complex reasoning capabilities of Multimodal Large Language Models (MLLMs)<n>In this paper, we propose the textbfGuided Verifier framework to address these structural limitations.<n>We develop a specialized data synthesis pipeline targeting multimodal hallucinations, constructing textbfCoRe dataset of process-level negatives and textbfCorrect-guide textbfReasoning trajectories to train the guided verifier.
  arXiv  Detail & Related papers  (2026-02-04T07:38:42Z)
• FIRE: Multi-fidelity Regression with Distribution-conditioned In-context Learning using Tabular Foundation Models [3.8824066002669855]
  Multi-fidelity (MF) regression often operates in regimes of extreme data imbalance.<n>We introduce FIRE, a training-free MF framework.<n>Fire delivers a stronger performance-time trade-off than seven state-of-the-art GP-based or deep learning MF regression methods.
  arXiv  Detail & Related papers  (2026-01-29T22:29:58Z)
• LANPO: Bootstrapping Language and Numerical Feedback for Reinforcement Learning in LLMs [73.27182315028021]
  LANPO is a framework that cleanly separates the roles of feedback: language guides exploration, while numerical rewards drive optimization.<n>Our work provides a robust method for integrating historical experiences into the LLM RL loop, creating more effective and data-efficient learning agents.
  arXiv  Detail & Related papers  (2025-10-18T15:51:19Z)
• DeLeaker: Dynamic Inference-Time Reweighting For Semantic Leakage Mitigation in Text-to-Image Models [55.30555646945055]
  Text-to-Image (T2I) models are vulnerable to semantic leakage.<n>We introduce DeLeaker, a lightweight approach that mitigates leakage by directly intervening on the model's attention maps.<n>SLIM is the first dataset dedicated to semantic leakage.
  arXiv  Detail & Related papers  (2025-10-16T17:39:21Z)
• Alignment Tipping Process: How Self-Evolution Pushes LLM Agents Off the Rails [103.05296856071931]
  We identify the Alignment Tipping Process (ATP), a critical post-deployment risk unique to self-evolving Large Language Model (LLM) agents.<n>ATP arises when continual interaction drives agents to abandon alignment constraints established during training in favor of reinforced, self-interested strategies.<n>Our experiments show that alignment benefits erode rapidly under self-evolution, with initially aligned models converging toward unaligned states.
  arXiv  Detail & Related papers  (2025-10-06T14:48:39Z)
• AdvChain: Adversarial Chain-of-Thought Tuning for Robust Safety Alignment of Large Reasoning Models [62.70575022567081]
  We propose AdvChain, an alignment paradigm that teaches models dynamic self-correction through adversarial CoT tuning.<n>Our work establishes a new direction for building more robust and reliable reasoning models.
  arXiv  Detail & Related papers  (2025-09-29T04:27:23Z)
• Towards Foundation Models for Zero-Shot Time Series Anomaly Detection: Leveraging Synthetic Data and Relative Context Discrepancy [33.68487894996624]
  Time series anomaly detection (TSAD) is a critical task, but developing models that generalize to unseen data remains a major challenge.<n>We introduce textttTimeRCD, a novel foundation model for TSAD built upon a new pre-training paradigm: Relative Context Discrepancy (RCD)<n>We show that textttTimeRCD significantly outperforms existing general-purpose and anomaly-specific foundation models in zero-shot TSAD.
  arXiv  Detail & Related papers  (2025-09-25T14:05:15Z)
• STARec: An Efficient Agent Framework for Recommender Systems via Autonomous Deliberate Reasoning [54.28691219536054]
  We introduce STARec, a slow-thinking augmented agent framework that endows recommender systems with autonomous deliberative reasoning capabilities.<n>We develop anchored reinforcement training - a two-stage paradigm combining structured knowledge distillation from advanced reasoning models with preference-aligned reward shaping.<n>Experiments on MovieLens 1M and Amazon CDs benchmarks demonstrate that STARec achieves substantial performance gains compared with state-of-the-art baselines.
  arXiv  Detail & Related papers  (2025-08-26T08:47:58Z)
• Beyond 'Aha!': Toward Systematic Meta-Abilities Alignment in Large Reasoning Models [86.88657425848547]
  Large reasoning models (LRMs) already possess a latent capacity for long chain-of-thought reasoning.<n>We explicitly align models with three meta-abilities: deduction, induction, and abduction, using automatically generated, self-verifiable tasks.<n>Our three stage-pipeline individual alignment, parameter-space merging, and domain-specific reinforcement learning, boosts performance by over 10% relative to instruction-tuned baselines.
  arXiv  Detail & Related papers  (2025-05-15T17:58:33Z)

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.