arXiv Computation and Language September 23, 2026 By Edward Xi Yang (Ertas AI)

Retrieved-Span Training for Efficient Query-Focused Meeting Summarization on QMSum

Read the original on arXiv Computation and Language →

The paper introduces a retrieved‑span training approach for query‑focused meeting summarization on the QMSum benchmark. By fine‑tuning a 406 M Fusion‑in‑Decoder model on 2,000‑word retrieved spans, the authors recover a 6.30 ROUGE‑1 loss incurred when moving from capped long input and achieve a test score of 36.33 ROUGE‑1, comparable to a larger 1.2 B system. The smaller model uses roughly one‑third the parameters and less than half the peak inference memory, while span‑regime fine‑tuning adds significant gains over the baseline. "whyItMatters":"The study demonstrates that efficient, smaller models can match or exceed larger systems on QMSum using span‑based fine‑tuning, offering a practical path for scalable query‑focused meeting summarization."

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

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 Computation and Language

Sep 22

BudgetMem: Training-Free Selective Memory for Cost-Efficient Long-Context Processing in Language Models

arXiv:2511. 04919v3 Announce Type: replace Abstract: Processing long documents with large language models (LLMs) is expensive: a single query over a 100K-token document can cost from tens of cents to over a dollar in API fees, depending on the model, and memory grows linearly with context length.

By Chandra Vamsi Krishna Alla, Harish Naidu Gaddam, Manohar Kommi, Sheikh Nazib Ahmed

llms nlp benchmarks

arXiv AI

Jun 9

Decision-Aware Memory Cards: Counterfactual-Inspired Context Selection and Compression for Tool-Using LLM Agents

arXiv:2606. 08151v1 Announce Type: new Abstract: Tool-using LLM agents often fail not because relevant text is absent, but because decisive evidence is not selected, compressed, or surfaced at action time.

By Xinyu Guan, Qianyang Zhao, Yuming Deng

llms rag agents fine-tuning benchmarks

arXiv Machine Learning

Jun 30

MemDelta: Controlled Baselines and Hidden Confounds in Agent Memory Evaluation

arXiv:2606. 29914v1 Announce Type: cross Abstract: Agent memory systems are increasingly evaluated against RAG and full-context baselines, but reported gains often mix changes in the memory method with changes in the language model, embedding model, or retrieval pipeline, making it unclear what is actually being measured.

By Kuan Wang

llms rag agents

arXiv AI

Jul 21

Beyond Memory Leaderboards: Evaluating Scientific Memory as Budgeted Context Restoration

arXiv:2607. 16848v1 Announce Type: cross Abstract: Long-term memory is becoming a core component of LLM agents, but most memory benchmarks evaluate conversations or compact summaries, while research agents need to restore evidence from full scientific papers.

By Maksim Sheverev, David Finkelstein, Sergey Nikolenko

llms rag agents benchmarks

arXiv Machine Learning

Sep 18

Intrinsic Sequence-Likelihood Confidence in Retrieval-Dominated Extractive QA: Two Pre-Specified Negatives, and What They Do and Do Not Attribute

The paper investigates whether confidence signals from fine‑tuned large language models can improve extractive question answering that relies heavily on retrieval. Experiments on four 7‑9B model families show that retrieval alone recovers 92–99.8% of the best possible accuracy, leaving little room for confidence‑based routing or adaptation to help. The sequence‑likelihood confidence metric, even after recalibration or temperature scaling, fails to provide a statistically significant benefit across different correctness criteria and answer lengths, and the study ultimately offers a set of pre‑specified negatives with explicit dependencies as its main contribution.

By Gunwoo Lee, Changmin Sung, Sang-Hwan Gwak, Ina Kim, Ji-Young Choi, Kyong-Ha Lee

llms nlp fine-tuning efficiency

One email a morning, machine-written

Related stories

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

BudgetMem: Training-Free Selective Memory for Cost-Efficient Long-Context Processing in Language Models

Decision-Aware Memory Cards: Counterfactual-Inspired Context Selection and Compression for Tool-Using LLM Agents

MemDelta: Controlled Baselines and Hidden Confounds in Agent Memory Evaluation

Beyond Memory Leaderboards: Evaluating Scientific Memory as Budgeted Context Restoration

Intrinsic Sequence-Likelihood Confidence in Retrieval-Dominated Extractive QA: Two Pre-Specified Negatives, and What They Do and Do Not Attribute