← Back to all news
arXiv AI By Hita Kambhamettu, Alyssa Hwang, Philippe Laban, Andrew Head

Attribution Gradients: Incrementally Unfolding Citations for Critical Examination of Attributed AI Answers

Read the original on arXiv AI →

The paper introduces attribution gradients, a method to enhance AI-generated answers by consolidating evidence counts, supporting or contradicting excerpts, source links, and contextual explanations into a single view. It also enables the unraveling of second‑degree citations directly within the answer. A lab study shows that using the full gradient improves readers’ critical engagement and deepens their understanding compared to standard citation and document QA designs.

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 AI.

Related stories

arXiv AI
Aug 20

DeepWeaver: Bridging the Evidence Synthesis Gap in Open-Ended Question Answering

DeepWeaver is a framework designed to improve open‑ended question answering by weaving noisy retrieved evidence into comprehensive, well‑cited answers. It introduces Thought Block Chains (TBCs) that organize claims, key information, and supporting evidence, and uses subordinate TBCs to refine and expand the evidence before final generation. Evaluations on LoQA and DeepResearch Bench show that DeepWeaver enhances content sufficiency, citation grounding, and detail preservation across multiple LLMs.

By Xujia Wang, Yizhe Zhang, Bin Xu, Lei Hou, Juanzi Li
llmsnlpbenchmarks
More like this →
Hugging Face Trending Papers
Aug 19

DeepWeaver: Bridging the Evidence Synthesis Gap in Open-Ended Question Answering

DeepWeaver addresses the evidence synthesis gap in open‑ended question answering by weaving noisy retrieved evidence into comprehensive answers. It introduces Thought Block Chains (TBCs) that organize claims, key information, and citations, allowing the system to revise and expand evidence before final generation. Evaluations on LoQA and DeepResearch Bench show improved content sufficiency, citation grounding, and detail preservation across multiple LLMs.

llmsnlpbenchmarks
More like this →
arXiv Computation and Language
Oct 9

CiteVQA: Benchmarking Evidence Attribution for Trustworthy Document Intelligence

CiteVQA is a benchmark that evaluates multimodal large language models on both answering document questions and providing element‑level bounding‑box citations for the evidence used. It contains 1,897 questions across 711 PDFs in seven domains and two languages, with ground‑truth citations generated via an automated pipeline and strict quality control. The evaluation metric, Strict Attributed Accuracy (SAA), rewards a prediction only when both the answer and the cited region are correct, revealing widespread attribution hallucination in current models.

By Dongsheng Ma, Jiayu Li, Zhengren Wang, Yijie Wang, Jiahao Kong, Weijun Zeng, Jutao Xiao, Jie Yang, Bangrui Xu, Yuhan Wang, Bin Wang, Conghui He
llmsmultimodalbenchmarkssafety
More like this →