Combating Instruction Conflict via Energy-Driven Latent Conflict Detection

• URL: http://arxiv.org/abs/2609.08646v1
• Date: Tue, 08 Sep 2026 12:14:43 GMT
• Title: Combating Instruction Conflict via Energy-Driven Latent Conflict Detection
• Abstract summary: Large Language Models (LLMs) are increasingly deployed with hierarchical instructions, yet they remain vulnerable to conflicts in which user directives override system-level constraints.<n>We introduce ELCD, a response-level latent conflict detector for post-generation, pre-delivery verification.
• Score: 6.701569487024643
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large Language Models (LLMs) are increasingly deployed with hierarchical instructions, yet they remain vulnerable to conflicts in which user directives override system-level constraints. Existing defense mechanisms predominantly focus on static input inspection and therefore fail to detect Response Drift, a phenomenon in which the model's final response violates system-level constraints despite seemingly compliant inputs. To bridge this gap, we introduce ELCD, a response-level latent conflict detector for post-generation, pre-delivery verification. Given the full generated output, ELCD constructs a composite hidden-state representation by concatenating the final-token embedding with the mean-pooled response embedding. It then optimizes a pairwise margin ranking objective to separate compliant and drifting responses in latent space. Extensive experiments across five mainstream LLMs ranging from 1.5B to 14B parameters demonstrate that ELCD significantly outperforms competitive baselines. Notably, it improves the PR-AUC on Llama-2-7B by approximately 30 percentage points and reduces the False Positive Rate at 95% TPR (FPR95) on Mistral-7B to 2.67%. These results suggest that ELCD provides a promising approach for latent instruction-conflict detection in open-weight or self-hosted LLM deployments.

Related papers

• ConflictRAG: Detecting and Resolving Knowledge Conflicts in Retrieval Augmented Generation [10.475475632969799]
  We present ConflictRAG, a conflict-aware RAG framework that detects, classifies, and resolves knowledge conflicts prior to answer generation.<n>The framework reduces API costs by 62% while maintaining 90.8% detection accuracy.
  arXiv  Detail & Related papers  (2026-05-17T07:25:29Z)
• Hierarchical Mixture-of-Experts with Two-Stage Optimization [84.70724165894501]
  We propose Hi-MoE, a grouped MoE framework that decomposes routing control into two coupled levels.<n>We observe consistent improvements over recent sparse-routing and grouped-MoE baselines across NLP and vision benchmarks.<n>In large-scale pre-training on 58B tokens, Hi-MoE-7B achieves a 5.6% perplexity reduction and a 40% improvement in expert balance over OLMoE-7B.
  arXiv  Detail & Related papers  (2026-05-08T09:21:46Z)
• BITS Pilani at SemEval-2026 Task 9: Structured Supervised Fine-Tuning with DPO Refinement for Polarization Detection [2.2588605422113606]
  We present a two-stage approach for detecting political polarization in social media text.<n>We fine-tune Qwen 2.5-7B-Instruct with LoRA using an interpretable slot-filling template.<n>Experiments on the SemEval 2026 POLAR shared task dataset show that preference-based refinement improves both accuracy and decreases false negatives without extra annotation.
  arXiv  Detail & Related papers  (2026-04-13T07:35:17Z)
• Improving Search Agent with One Line of Code [68.58667107354253]
  Tool-based Agentic Reinforcement Learning (TARL) has emerged as a promising paradigm for training search agents to interact with external tools.<n>We propose textbfSearch textbfAgent textbfPolicy textbfOptimization (textbfSAPO), which stabilizes training via a conditional token-level KL constraint.
  arXiv  Detail & Related papers  (2026-03-10T04:07:39Z)
• Hybrid Responsible AI-Stochastic Approach for SLA Compliance in Multivendor 6G Networks [1.8219577154655007]
  This paper presents a hybrid responsible AI-stochastic learning framework that embeds fairness, robustness, and auditability directly into the network control loop.<n>An RAAP continuously records AI-driven decision trajectories and produces user-level SLA summaries and operator-level responsibility analytics.<n> Experimental evaluations on synthetic two-class multigroup datasets demonstrate that the proposed hybrid model improves the accuracy of the worst group by up to 10.5%.
  arXiv  Detail & Related papers  (2026-02-10T14:45:59Z)
• CARE What Fails: Contrastive Anchored-REflection for Verifiable Multimodal [84.71254539482369]
  Group-relative reinforcement learning with verifiable rewards (RLVR) often wastes the most informative data it already has the failures.<n>We present CARE, a failure-centric post-training framework for multimodal reasoning that turns errors into supervision.<n> CARE improves accuracy and training smoothness while explicitly increasing the share of learning signal that comes from failures.
  arXiv  Detail & Related papers  (2025-12-22T16:34:21Z)
• Enhancing the Outcome Reward-based RL Training of MLLMs with Self-Consistency Sampling [90.87033586963828]
  Outcome-reward reinforcement learning (RL) is a common and increasingly significant way to refine the step-by-step reasoning of multimodal large language models (MLLMs)<n>We propose Self-Consistency Sampling (SCS) to correct this issue.<n>Based on Qwen2.5-VL-7B-Instruct, SCS improves accuracy by up to 7.7 percentage points on six multimodal benchmarks with negligible extra computation.
  arXiv  Detail & Related papers  (2025-11-13T18:59:57Z)
• RaCoT: Plug-and-Play Contrastive Example Generation Mechanism for Enhanced LLM Reasoning Reliability [12.67288560758937]
  We propose RaCoT (Retrieval-aware Contrastive-of-Thought), a novel framework that shifts contrastive thinking to the pre-retrieval stage.<n>RaCoT guides the model to proactively focus on the critical details that determine answer divergence"
  arXiv  Detail & Related papers  (2025-10-26T15:06:44Z)
• Causally-Enhanced Reinforcement Policy Optimization [36.523007244998695]
  Causally-Enhanced Policy Optimization (CE-PO) is a drop-in reward-shaping framework that augments policy optimization with a differentiable proxy for causal coherence.<n>CE-PO estimates model-internal influence with Jacobian-based sensitivities, counterfactually hardens these signals to suppress nuisance cues, and fuses the resulting coherence score with task-accuracy feedback.<n> Experimental results across 4 datasets show that CE-PO improves accuracy over baselines by 5.49% on average (up to 9.58%), while improving robustness to correlation-causation flips and light counterfactual edits.
  arXiv  Detail & Related papers  (2025-09-27T04:10:16Z)
• GRPO-CARE: Consistency-Aware Reinforcement Learning for Multimodal Reasoning [53.894789613838654]
  We introduce SEED-Bench-R1, a benchmark with complex real-world videos requiring balanced perception and reasoning.<n>Using SEED-Bench-R1, we find that standard GRPO, while improving answer accuracy, often reduces logical coherence between reasoning steps and answers, with only a 57.9% consistency rate.<n>We propose GRPO-CARE, a consistency-aware RL framework optimizing both answer correctness and reasoning coherence without explicit supervision.
  arXiv  Detail & Related papers  (2025-06-19T08:49:13Z)
• Retrieval is Not Enough: Enhancing RAG Reasoning through Test-Time Critique and Optimization [58.390885294401066]
  Retrieval-augmented generation (RAG) has become a widely adopted paradigm for enabling knowledge-grounded large language models (LLMs)<n>RAG pipelines often fail to ensure that model reasoning remains consistent with the evidence retrieved, leading to factual inconsistencies or unsupported conclusions.<n>We propose AlignRAG, a novel iterative framework grounded in Critique-Driven Alignment (CDA)<n>We introduce AlignRAG-auto, an autonomous variant that dynamically terminates refinement, removing the need to pre-specify the number of critique iterations.
  arXiv  Detail & Related papers  (2025-04-21T04:56:47Z)
• Chain of Evidences and Evidence to Generate: Prompting for Context Grounded and Retrieval Augmented Reasoning [3.117335706912261]
  Chain of Evidences (CoE) and Evidence to Generate (E2G) are built upon two unique strategies.<n>Instead of unverified reasoning claims, our innovative approaches leverage the power of "evidence for decision making"<n>Our framework consistently achieves remarkable results across various knowledge-intensive reasoning and generation tasks.
  arXiv  Detail & Related papers  (2024-01-11T09:49:15Z)

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.