arXiv Machine Learning August 6, 2026 By Zhengpei Hu, Kai Li, Dapeng Fu, Xuechao Zou, Yuanhao Tang, Yue Li, Tengfei Cao, Jianqiang Huang

Relevant but Incomplete: Referential Dangling as a Paradigm-Level Failure Mode in Hard Prompt Compression

Read the original on arXiv Machine Learning →

arXiv:2608. 04569v1 Announce Type: cross Abstract: Hard prompt compression reduces long-context inference cost by independently scoring tokens, sentences, or chunks and retaining the highest-scoring units under a budget.

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

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

Jun 24

CAVEWOMAN: How Large Language Models Behave Under Linguistic Input and Output Compression

arXiv:2606. 24083v1 Announce Type: cross Abstract: "Talk short.

By Morayo Danielle Adeyemi, Ryan A. Rossi, Franck Dernoncourt

llms benchmarks

arXiv Machine Learning

Jul 1

InfoFlow KV: Information-Flow-Aware KV Recomputation for Long Context

arXiv:2603. 05353v2 Announce Type: replace Abstract: Retrieval-augmented generation (RAG) for long-context question answering is bottlenecked by inference-time prefilling over large retrieved contexts.

By Xin Teng, Canyu Zhang, Shaoyi Zheng, Danyang Zhuo, Tianyi Zhou, Shenji Wan

llms rag nlp multimodal benchmarks

arXiv AI

Jun 15

Sentinel: Decoding Context Utilization via Attention Probing for Efficient LLM Context Compression

arXiv:2505. 23277v3 Announce Type: replace-cross Abstract: Retrieval-augmented generation (RAG) often suffers from long and noisy retrieved contexts.

By Yong Zhang, Heng Li, Yanwen Huang, Ning Cheng, Yang Guo, Yun Zhu, Yanmeng Wang, Shaojun Wang, Jing Xiao

llms rag

arXiv AI

Jun 15

Verbatim Chunks Beat Extracted Artifacts: A Controlled Ablation of Memory Representations for Long LLM Conversations

arXiv:2601. 00821v3 Announce Type: replace Abstract: A growing class of conversational-memory systems compresses dialogue history into structured artifacts -- extracted facts, decisions, or events -- on the premise that distilled structure retrieves better than raw text.

By Tao An

llms rag efficiency benchmarks

arXiv Machine Learning

Jul 13

Memory-Managed Long-Context Attention: Bounded Editable Memory with a Hard Lifecycle and Calibrated Sparse Fallback

arXiv:2606. 28876v2 Announce Type: replace-cross Abstract: We study memory-managed long-context attention: explicit bounded memory with a learned query-independent writer, lifecycle control, query-aware reading, calibrated sparse fallback, and frozen-LLM generation from raw evidence.

By Junyi Zou, Avrova Donz

llms

Related stories

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

CAVEWOMAN: How Large Language Models Behave Under Linguistic Input and Output Compression

InfoFlow KV: Information-Flow-Aware KV Recomputation for Long Context

Sentinel: Decoding Context Utilization via Attention Probing for Efficient LLM Context Compression

Verbatim Chunks Beat Extracted Artifacts: A Controlled Ablation of Memory Representations for Long LLM Conversations

Memory-Managed Long-Context Attention: Bounded Editable Memory with a Hard Lifecycle and Calibrated Sparse Fallback