aiqbench analysis
The proposed CCHD framework improves hallucination detection by integrating paraphrase-consistency constraints into the training objective, eliminating the need for expensive data synthesis or inference-time overhead. By leveraging gradient descent-ascent with Lagrange multipliers, this method effectively bounds divergence across semantically equivalent views while maintaining label fidelity. The approach demonstrates superior performance on standard factuality benchmarks using DeBERTa and Flan-T5 backbones, offering a scalable solution for addressing LLM factuality issues.
Constrained Paraphrase Consistency for LLM Hallucination Detection
arXiv:2606.08158v1 Announce Type: new Abstract: Large language models (LLMs) can generate factually inconsistent claims, motivating accurate and scalable hallucination detectors. Prior work largely enlarges training sets via synthesis or new annotations, introducing increasing cost and potential bias while underusing the consistency implied by semantically equivalent paraphrases. We propose Consistency-Constrained Hallucination Detector (CCHD), which formulates training as a constrained optimization problem. The standard cross-entropy on original document-claim pairs is complemented by (i) paraphrase-consistency constraints bounding divergence across paraphrased views, and (ii) label-preservation constraints tying paraphrases to ground truth. We solve the problem by gradient descent-ascent over model parameters and per-view Lagrange multipliers, adding only a few scalar dual variables and no inference-time overhead. With DeBERTa and Flan-T5 backbones, CCHD consistently outperforms strong baselines (FactCG, MiniCheck, and AlignScore) on standard factuality benchmarks, demonstrating its superiority on hallucination detection.
Abstract:Large language models (LLMs) can generate factually inconsistent claims, motivating accurate and scalable hallucination detectors. Prior work largely enlarges training sets via synthesis or new annotations, introducing increasing cost and potential bias while underusing the consistency implied by semantically equivalent paraphrases. We propose Consistency-Constrained Hallucination Detector (CCHD), which formulates training as a constrained optimization problem. The standard cross-entropy on original document-claim pairs is complemented by (i) paraphrase-consistency constraints bounding divergence across paraphrased views, and (ii) label-preservation constraints tying paraphrases to ground truth. We solve the problem by gradient descent-ascent over model parameters and per-view Lagrange multipliers, adding only a few scalar dual variables and no inference-time overhead. With DeBERTa and Flan-T5 backbones, CCHD consistently outperforms strong baselines (FactCG, MiniCheck, and AlignScore) on standard factuality benchmarks, demonstrating its superiority on hallucination detection.
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
arXiv:2606.08158 [cs.CL]
(or arXiv:2606.08158v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2606.08158 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 229 words
Related coverage
Latent reasoning architectures would likely undermine CoT, our strongest oversight tool
AI Alignment Forum · Sep 23, 2026
[AINews] not much happened today
Latent Space · Sep 18, 2026
BEACON: Behavioral Entropy Aggregation for Cross-Model Hallucination Detection in Large Language Models
arXiv · Jun 9, 2026
[AINews] High Taste Testers Ep. 1
Latent Space · Oct 17, 2026