arXiv Machine Learning June 5, 2026 By Govind Ramesh, Yao Dou, Wei Xu

Localizing Prompt Ambiguity in Large Language Models with Probe-Targeted Attribution

Read the original on arXiv Machine Learning →

arXiv:2606. 05486v1 Announce Type: cross Abstract: Prompt ambiguity is a common source of failure in large language models, but is difficult to localize because it is a latent property of the prompt, while existing attribution methods are designed to explain observable outputs such as logits or generated tokens.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.

arXiv Machine Learning

Jul 7

The Truncation Blind Spot: How Decoding Strategies Systematically Exclude Human-Like Token Choices

arXiv:2603. 18482v2 Announce Type: replace-cross Abstract: Standard decoding strategies for text generation, including top-$k$, nucleus sampling, and contrastive search, select tokens based on likelihood, restricting outputs to high-probability regions.

By Esteban Garces Arias, Nurzhan Sapargali, Christian Heumann, Matthias A{\ss}enmacher

llms safety

arXiv Machine Learning

Jun 16

Emergent retokenization symmetry in large language models: phenomenology and applications

arXiv:2606. 15521v1 Announce Type: cross Abstract: Tokenization introduces representational redundancy: under a fixed token vocabulary, every byte string admits many valid token encodings, or segmentations, that decode to the same surface string.

By Kanishk Jain, Matthew Day, Tankut Can

llms computer-vision nlp benchmarks

arXiv AI

Jul 7

Detecting Hallucinations in Retrieval-Augmented Generation through Grounding-Aware Sensitivity by Perturbation (GASP)

arXiv:2607. 04223v1 Announce Type: cross Abstract: Retrieval-augmented generation (RAG) reduces but does not eliminate hallucination, and existing detectors return a single answer-level score that does not indicate which sentence is unsupported, or why.

By Mohamed Aly Bouke

rag nlp benchmarks safety

arXiv AI

Jul 13

GrAInS: Gradient-based Attribution for Inference-Time Steering of LLMs and VLMs

arXiv:2507. 18043v2 Announce Type: replace-cross Abstract: Inference-time steering methods offer a lightweight alternative to fine-tuning large language models (LLMs) and vision-language models (VLMs) by modifying internal activations at test time without updating model weights.

By Duy Nguyen, Archiki Prasad, Elias Stengel-Eskin, Mohit Bansal

llms fine-tuning multimodal safety

arXiv AI

Jul 7

Consistent but Miscalibrated: Evaluating LLM Limitations for Risk Communication in Natural Language

arXiv:2607. 03882v1 Announce Type: cross Abstract: LLMs are increasingly deployed as post-hoc explainers of AI-generated outputs, yet it remains unclear whether they can reliably communicate probabilistic information in natural language.

By Diego Cerda-Mardini, Sarath Chandar, Sreenath Madathil

llms

arXiv Machine Learning

Aug 12

The Parser Already Knows: Lightweight Bias Correction in Constrained Decoding

arXiv:2608. 10137v1 Announce Type: cross Abstract: Grammar Constrained Decoding (GCD) forces Language Models (LMs) to produce syntactically valid outputs by masking out non-conforming tokens at each step.

By I\c{s}{\i}l \"Ozg\"u, Yaoxuan Wu, Guy Van den Broeck, Miryung Kim

llms efficiency safety

Related stories

The Truncation Blind Spot: How Decoding Strategies Systematically Exclude Human-Like Token Choices

Emergent retokenization symmetry in large language models: phenomenology and applications

Detecting Hallucinations in Retrieval-Augmented Generation through Grounding-Aware Sensitivity by Perturbation (GASP)

GrAInS: Gradient-based Attribution for Inference-Time Steering of LLMs and VLMs

Consistent but Miscalibrated: Evaluating LLM Limitations for Risk Communication in Natural Language

The Parser Already Knows: Lightweight Bias Correction in Constrained Decoding