arXiv Computation and Language

Detect, Remask, Repair: Diffusion Editing for Faithful Summarization of Evolving Contexts

The paper introduces DETECT-REMASK-REPAIR, a diffusion-based method for updating outdated spans in existing summaries while keeping supported content intact. It identifies, masks, and repairs only the changed regions using masked diffusion language models. Experiments on DialogSum and a new StreamSum benchmark show that this localized repair improves faithfulness, reduces repair time to under half a second, and offers trade‑offs between faithfulness, speed, and preservation of the original summary.

arXiv Computation and Language
Sep 3

CARPAS: Towards Content-Aware Refinement of Provided Aspects for Summarization in Large Language Models

The paper introduces CARPAS, a new task that dynamically refines user-provided aspects for aspect-based summarization in large language models (LLMs). It presents three new datasets and evaluates four prompting strategies, finding that LLMs tend to over-generate aspects, leading to overly long and misaligned summaries. To address this, the authors propose a two-stage framework that first generates lightweight scope guidance before aspect refinement and summarization, which improves focus, reduces over-generation, and enhances performance across all datasets.

By Yong-En Tian, Yu-Chien Tang, An-Zi Yen, Wen-Chih Peng
arXiv AI
Jul 14

PRISM Edit: One Vector for All Temporal Answers

arXiv:2607. 11327v1 Announce Type: cross Abstract: Model editing keeps large language models (LLMs) up to date without retraining, but temporal facts expose a limitation of the prevailing locate-and-edit paradigm: an update is not always a replacement.

By Chen Huang (Tsinghua University), Qi Zheng (Tsinghua University), Ruiqin Zheng (ByteDance), Long Zeng (Tsinghua University), Yuantong Xu (ByteDance)
arXiv AI
Aug 17

EchoChange: A Diffusion Language Model with Dual Pass Remasking for Factual Remote Sensing Disaster Change Captioning

arXiv:2608. 01856v2 Announce Type: replace Abstract: Bi-temporal remote-sensing disaster change captioning often needs to identify sparse and spatially localized changes across large pre- and post-event scenes and then translate them into coherent, factual descriptions.

By Dongwei Sun, Bowen Yao, Yujie Zhang, Pei Liu, Jing Yao, Xiangyong Cao
arXiv Computation and Language
Aug 28

Dependency-Aware Revocable Decoding for Efficient Diffusion Large Language Model Inference

The paper introduces Dependency-Aware Revocable Decoding (DARD), a training‑free framework for diffusion large language models that separates tokens into masked, candidate, and unmasked states. DARD verifies candidate tokens using a selective context that excludes less reliable tokens and adaptively regulates their influence on subsequent decoding. Experiments on 12 textual and multimodal benchmarks across three open‑source dLLMs show that DARD improves the speed‑quality Pareto frontier, achieving a 2.71× speedup and a 4.35‑point CIDEr gain over Saber on Flickr30K.

By Wooje Park, Insu Lee, Minyoung Noh, Jaeyun Jang, Sungmin Lee, Kyuhong Shim, Byonghyo Shim
arXiv Computation and Language
Sep 11

Recognizing Is Not Reversing: A Controlled Inversion Test of Fact-Preserving News Framing

The paper introduces a controlled inversion test to evaluate whether large language models can reverse known framing transformations in news articles while preserving facts. Using 60 articles and three framing types—evaluative lexis, agency realization, and information salience—the study generates 540 paired variants. Results show high factual preservation (~0.84) but low reversal success (0.044–0.068), indicating that recognizing a framing does not guarantee its undoing.

By Yi Liu
arXiv Machine Learning
Aug 20

BERTilda: Explainable Topic Lifecycle Tracking with Split/Merge Detection via Similarity-and-Flow Temporal Graphs

BERTilda is an explainable framework for tracking topic lifecycles in longitudinal text streams. It discovers topics independently in each time window using an embedding‑based topic model, then links topics across adjacent windows via a temporal graph that uses both semantic similarity and a bidirectional coverage signal derived from tweet‑to‑topic attribution. The graph‑based rules identify continuations, splits, merges, disappearances, and unclear transitions, and the method achieves up to 87% agreement with human annotators on a gold‑standard subset.

By Cl\'audia Oliveira, \'Alvaro Figueira