5arXiv cs.CL (Computation and Language)·
CM-LRS: A capital markets reliability benchmark for LLM workflow outputs
Researchers introduce CM-LRS (Capital Markets LLM Reliability Score), a seven-dimension evaluation framework assessing LLM outputs at the workflow level rather than the question-answer layer, targeting regulated capital-markets use cases such as DCM/ECM term extraction, M&A comparables, and issuer profiling. The benchmark is demonstrated on five workflows using public SEC EDGAR and UK takeover filings, scoring four models across four LLM judges. Key findings: frontier closed-source models cluster tightly (Sonnet 4.6 = 4.31, Opus 4.7 = 4.30, GPT-5.5 = 4.09) while Llama 3.3 70B lags at 3.15, with the gap concentrated in retrieval and synthesis tasks rather than extraction. The work advances domain-specific evaluation methodology for high-stakes financial workflows where regulatory defensibility matters.
Related guides (4)
Enterprise Deployment PatternsTopic guide
Enterprise Deployment Patterns: How Organizations Are Moving LLMs from Demo to Production
Read asIn-depth
Related events (8)
DECODEM: Benchmark for LLM extraction of corporate governance variables from legal documents
DECODEM introduces benchmark datasets pairing corporate charters and bylaws with human annotations for evaluating automated extraction of governance variables using LLMs. The paper evaluates multiple extraction pipelines varying in prompt design, task decomposition, and document handling, finding frontier models achieve high accuracy on most provisions. A key finding is that elaborate prompting and cascading pipelines do not consistently improve frontier model performance but substantially narrow the gap with smaller efficiency-oriented models, suggesting pipeline design can partially substitute for model capability in structured legal document extraction.
NCRE-based benchmark reveals frontier LLMs top out at 68.8% on professional Office automation tasks
Researchers introduce an evaluation suite derived from China's National Computer Rank Examination (NCRE), comprising 200 practical tasks across Word, Excel, and PowerPoint scored via 7,118 machine-gradable criteria. Seven frontier LLMs are benchmarked: single-turn models peak at 36.6% Score Rate, while a full agentic system with execution feedback and iterative repair reaches 68.8%, still well below the 95.5% community-reference score. The results demonstrate that fine-grained, long-horizon Office document automation remains a significant unsolved challenge for current LLM and agent systems despite strong code-generation capabilities.
Paper challenges LLM expert-level claims by measuring variance and error magnitude in code-based data analysis tasks
A new arXiv paper argues that standard LLM benchmarks overstate model capabilities by focusing on average performance on training-data-adjacent tasks while ignoring response variance and error magnitude. The authors introduce a novel benchmark requiring frontier LLMs to write code for data analysis tasks, comparing results against human expert submissions. Human experts outperformed the frontier LLM on average across multiple metrics and showed lower performance variability. The findings challenge the prevailing narrative that LLMs perform at human-expert level on knowledge economy tasks.
Meta-evaluation of LLM-generated rubrics for paper reproduction benchmarks
A new arXiv preprint presents the first systematic meta-evaluation of LLM-generated rubrics for assessing paper reproduction tasks, addressing scalability limitations of expert-constructed rubrics in benchmarks like PaperBench. The authors test four generation settings across two backbone models, evaluating rubric quality both intrinsically (semantic similarity) and extrinsically (score alignment with ground-truth). Results show augmented generation settings can approach human-baseline alignment, but LLM-generated rubrics tend to be overly fine-grained, score-biased, and domain-insensitive.
LLMs automate reproducibility assessments in social and behavioral sciences, outperforming human reanalysts
A preprint from arXiv demonstrates that an LLM pipeline can automate reproducibility assessments of published social and behavioral science studies, recovering original effect sizes in 41% of cases (vs. 34% for human reanalysts) and reaching the same qualitative conclusion in 96% of cases (vs. 74% for humans). The study evaluated 76 published studies with predefined claims. The results suggest LLMs could serve as a scalable tool for systematic auditing of empirical research, addressing the resource-intensive nature of traditional reproducibility efforts.
German Central Bank case study applies LLMs to securities prospectus eligibility verification
Researchers present the first case study applying LLMs to the Deutsche Bundesbank's securities collateral eligibility verification process, replacing traditional NER-based pipelines with a generative information extraction approach. The system decomposes the task into extraction, normalization, and interpretation stages, handling OCR noise and bilingual (German-English) content in lengthy financial prospectuses. Results show up to 91% precision in document-level eligibility decisions with a conservative false-acceptance profile. The paper also introduces an LLM-as-a-judge evaluation methodology for semantic assessment of extraction quality.
RRBench: evaluating open-weight LLMs on longitudinal data preparation tasks in governance-restricted settings
Researchers from UCL introduce RRBench, an open-source benchmark and evaluation framework for assessing open-weight LLMs on data preparation tasks in longitudinal population studies, where cloud-based models are prohibited by data governance rules. The benchmark uses ground-truth cleaning scripts from six sweeps of a British cohort study, covering tasks like category harmonization and multi-wave merging, with automated evaluation of LLM-generated R code. Current 31–35B parameter open-weight models achieve up to 87.9% average task completion on consumer-grade hardware, suggesting a viable path for AI-assisted data preparation in privacy-sensitive research contexts.
Benchmarking LLM judges for citation verification in deep-research systems
A new arXiv paper evaluates 8 LLM judges from 3 model families on citation quality assessment for deep-research systems, testing across 1,248 rubric decisions with human-reviewed gold labels. The study finds that cheaper models remain competitive with frontier models — GPT-5-mini achieves the strongest source-relevance F1 at 0.908 — but judges differ substantially in directional bias (pass-rate drift, false positive/negative rates) even when scalar F1 scores are similar. The key finding is that scalar F1 obscures biases that would be directly reinforced in an RL training loop, making judge calibration a prerequisite before using citation rubrics as reward signals.