Detection Without Correction: A Two-Parameter Decomposition of Multi-Stage LLM Pipelines

• URL: http://arxiv.org/abs/2605.27559v1
• Date: Tue, 26 May 2026 18:26:15 GMT
• Title: Detection Without Correction: A Two-Parameter Decomposition of Multi-Stage LLM Pipelines
• Abstract summary: We study multi-agent debate, intrinsic self-correction, or retrieval-augmented verification pipelines.<n>We find that the conditional miscorrection rate is consistently dominant.<n>We characterizes detection threshold as a stable model/protocol-level regularity.
• Score: 0.764671395172401
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multi-stage LLM pipelines that perform multi-agent debate, intrinsic self-correction, or retrieval-augmented verification exhibit puzzling aggregate behaviors: accuracy plateaus and reversals across rounds, non-replication of debate gains on contemporary frontier models, intrinsic self-correction degradation, and qualitative cross-provider divergence in debate dynamics. Downstream agent response can be operationalized as two coupled decisions: detection (whether to treat upstream content as authoritative) and conditional generation (what to produce if not). This decomposition yields four observable response regimes, of which detection-without-correction is the load-bearing failure mode. Across a nine-cell empirical grid spanning four model families, four benchmarks (GSM8K, MATH-500, GPQA-Diamond, AIME), and two methods (multi-agent debate, intrinsic self-correction), we find that the conditional miscorrection rate is consistently dominant (53-94% across cohorts) while detection rate varies contextually by more than an order of magnitude. The framework unifies the four phenomena above as signatures of a common mechanism and characterizes detection threshold as a stable model/protocol-level regularity that persists across methods at matched benchmark difficulty.

Related papers

• From Tokens to Semantics: Leveraging Complementary Signals for Hallucination Detection in Black-Box LLMs [0.9751538234290761]
  Missing fabrications can harm users and institutions, while false alarms consume limited human-review capacity.<n>We study two signals accessible through black-box model APIs: semantic entropy, and uncertainty derived from token log-probabilities.<n>We extend token-based uncertainty detection by aggregating token-level signals across sampled responses through our TopK method.
  arXiv  Detail & Related papers  (2026-09-02T14:50:24Z)
• ANCHOR: Autoregressive Non-intrusive Chunk-Ordered Refinement for Joint Multi-Resolution Speech Quality Modeling [62.983372215479925]
  Streaming and generative systems require incremental estimation from partial audio.<n>We propose ANCHOR, reformulating incremental assessment as a multi-resolution autore task.<n> Experiments show substantial robustness under partial input, including a 48% PLCMOS error reduction on 2-second prefixes.
  arXiv  Detail & Related papers  (2026-06-08T22:46:30Z)
• Cascading Hallucination in Agentic RAG: The CHARM Framework for Detection and Mitigation [0.0]
  We introduce CHARM, an architectural framework for detecting and interrupting error propagation in multi-step reasoning pipelines.<n> CHARM comprises four components - stage-level fact verification, cross-stage consistency tracking, confidence propagation monitoring, and cascade resolution triggering.<n>We evaluate CHARM on HotpotQA, MuSiQue, 2WikiMultiHopQA, and a custom adversarial dataset across LangChain agentic pipeline configurations.
  arXiv  Detail & Related papers  (2026-06-03T04:33:47Z)
• IV-ICL: Bounding Causal Effects with Instrumental Variables via In-Context Learning [11.804716294874018]
  IV-ICL is an amortized Bayesian in-context learning method that learns the marginal posterior distribution of the causal effects directly and derives bounds as its quantiles.<n>We evaluate IV-ICL on synthetic and semi-synthetic IV benchmarks and show it produces intervals that are more reliably valid and more informative.
  arXiv  Detail & Related papers  (2026-05-13T03:00:08Z)
• How Independent are Large Language Models? A Statistical Framework for Auditing Behavioral Entanglement and Reweighting Verifier Ensembles [46.63622714488747]
  Shared pretraining data, distillation, and alignment pipelines can induce hidden behavioral dependencies, latent entanglement.<n>In practice, this manifests as correlated reasoning patterns and synchronized failures.<n>We develop a statistical framework for auditing behavioral entanglement among black-box language models.
  arXiv  Detail & Related papers  (2026-04-08T23:32:06Z)
• TRIAGE: Type-Routed Interventions via Aleatoric-Epistemic Gated Estimation in Robotic Manipulation and Adaptive Perception -- Don't Treat All Uncertainty the Same [2.755751829139168]
  Most uncertainty-aware robotic systems collapse prediction uncertainty into a single scalar score and use it to trigger uniform corrective responses.<n>This aggregation obscures whether uncertainty arises from corrupted observations or from mismatch between the learned model and the true system dynamics.<n>We introduce a lightweight post hoc framework that decomposes uncertainty into aleatoric and epistemic components and uses these signals to regulate system responses at inference time.
  arXiv  Detail & Related papers  (2026-03-09T09:07:43Z)
• Search-R2: Enhancing Search-Integrated Reasoning via Actor-Refiner Collaboration [49.9937230730202]
  We propose Search-R2, a novel Actor-Refiner collaboration framework that enhances reasoning through targeted intervention.<n>Our approach decomposes the generation process into an Actor, which produces initial reasoning trajectories.<n>We show that Search-R2 consistently outperforms strong RAG and RL-based baselines across model scales.
  arXiv  Detail & Related papers  (2026-02-03T15:32:09Z)
• Unsupervised Anomaly Detection in Multi-Agent Trajectory Prediction via Transformer-Based Models [45.08545174556591]
  We propose an unsupervised anomaly detection framework based on a multi-agent Transformer.<n>A dual evaluation scheme has been proposed to assess both detection stability and physical alignment.<n>Our framework identifies 388 unique anomalies missed by Time-to-Collision and statistical baselines.
  arXiv  Detail & Related papers  (2026-01-28T08:33:10Z)
• Prompt Injection Evaluations: Refusal Boundary Instability and Artifact-Dependent Compliance in GPT-4-Series Models [0.0]
  We evaluate two models, GPT-4.1 and GPT-4o, using 3,274 perturbation runs derived from refusal-inducing prompt injection attempts.<n>We find that artifact type is a stronger predictor of refusal failure than perturbation style.<n>We conclude that refusal behavior is a probabilistic, artifact-dependent boundary phenomenon rather than a stable binary property.
  arXiv  Detail & Related papers  (2026-01-25T17:14:33Z)
• DVD: A Robust Method for Detecting Variant Contamination in Large Language Model Evaluation [24.086354908256293]
  textbfDVD is a single-sample detector that models the local output distribution induced by temperature sampling.<n>We construct the first benchmark for variant contamination across two domains Omni-MATH and SuperGPQA.<n>textbfDVD consistently outperforms perplexity-based, Min-$k$%++, edit-distance (CDD), and embedding-similarity baselines.
  arXiv  Detail & Related papers  (2026-01-08T12:48:40Z)
• 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)
• Identification of Causal Direction under an Arbitrary Number of Latent Confounders [54.76982125821112]
  In real-world scenarios, observed variables may be affected by multiple latent variables simultaneously.<n>We make use of the joint higher-order cumulant matrix of the observed variables constructed in a specific way.<n>We show that, surprisingly, causal asymmetry between two observed variables can be directly seen from the rank deficiency properties of such higher-order cumulant matrices.
  arXiv  Detail & Related papers  (2025-10-26T15:10:00Z)
• Eigen-1: Adaptive Multi-Agent Refinement with Monitor-Based RAG for Scientific Reasoning [53.45095336430027]
  We develop a unified framework that combines implicit retrieval and structured collaboration.<n>On Humanity's Last Exam (HLE) Bio/Chem Gold, our framework achieves 48.3% accuracy.<n>Results on SuperGPQA and TRQA confirm robustness across domains.
  arXiv  Detail & Related papers  (2025-09-25T14:05:55Z)
• Cross-Attention is Not Enough: Incongruity-Aware Dynamic Hierarchical Fusion for Multimodal Affect Recognition [69.32305810128994]
  Incongruity between modalities poses a challenge for multimodal fusion, especially in affect recognition. We propose the Hierarchical Crossmodal Transformer with Dynamic Modality Gating (HCT-DMG), a lightweight incongruity-aware model. HCT-DMG: 1) outperforms previous multimodal models with a reduced size of approximately 0.8M parameters; 2) recognizes hard samples where incongruity makes affect recognition difficult; 3) mitigates the incongruity at the latent level in crossmodal attention.
  arXiv  Detail & Related papers  (2023-05-23T01:24:15Z)
• Information Directed Sampling for Linear Partial Monitoring [112.05623123909895]
  We introduce information directed sampling (IDS) for partial monitoring with a linear reward and observation structure. IDS achieves adaptive worst-case regret rates that depend on precise observability conditions of the game. We extend our results to the contextual and the kernelized setting, which significantly increases the range of possible applications.
  arXiv  Detail & Related papers  (2020-02-25T21:30:56Z)

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.