arXiv AI
Source-Free MT Evaluation Is Not MT Evaluation
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv Computation and Language
Sep 15
In the Blind: Building Pseudo-References for MT Evaluation
arXiv:2609.13611v1 Announce Type: new Abstract: The WMT26 General MT task evaluates systems on 10 language pairs that have no human references (neither translated from scratch nor post-edited from MT...
arXiv AI
Sep 7
Attributable by Construction: Claim-Anchored Provenance for Multi-Document Summarization
The paper introduces CAMS, a Claim‑Anchored Multi‑Document Summarization framework that decomposes source documents into atomic claims, resolves provenance deterministically from verbatim quotes to token spans, clusters equivalent claims across documents, and rewrites summaries so each sentence ends with claim identifiers linking back to source spans. CAMS separates provenance (an invariant for each emitted sentence) from faithfulness (an objective encouraged by selection, rewriting, and verification). Evaluations on MultiNews, DiverseSumm, and zero‑shot WCEP show that CAMS matches strong baselines in summary quality while improving faithfulness and citation precision, raising attribution accuracy from 38% to 64% and reducing human verification time per claim by 3.4×.
arXiv AI
Jun 24
Faithful by Construction: Claim-Anchored Attribution for Multi-Document Summarization
arXiv:2606. 23989v1 Announce Type: cross Abstract: End-to-end large language models (LLMs) produce fluent multi-document summaries but remain prone to hallucination, and the attributions they offer are typically coarse (whole documents or passages) and generated post hoc, leaving each summary statement hard to verify.
arXiv AI
Oct 15
claimcheck: Measuring and Improving Specification Faithfulness
arXiv:2610.17358v1 Announce Type: cross Abstract: Formal verification proves that a program satisfies its specification, but says nothing about whether the specification captures what the author mean...
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.
arXiv Computation and Language
Oct 16
Reasoning without Gold Standards: A Proxy-Judge Theory of Autoformalization
The paper introduces a reference‑free proxy‑judge framework for autoformalization, separating correctness into grounding, statement faithfulness, coverage, and inference. It presents three theoretical results linking proxy error rates to pass‑but‑wrong rates, the impact of false rejections on refinement, and a decomposition of final pass rates. The framework is instantiated with eight proxies and evaluated on ProofNet, miniF2F, e‑SNLI, and ProntoQA, showing that the predictions hold and identifying failure routes.