LLMs copy biased analyst ratings in investment decisions but a new detection method encourages independent reasoning and can improve stock return predictions beyond human levels.
FinQA: A dataset of numerical reasoning over financial data
10 Pith papers cite this work, alongside 8 external citations. Polarity classification is still indexing.
10
Pith papers citing it
8
external citations · OpenAlex
citation-role summary
background 2
dataset 1
citation-polarity summary
representative citing papers
TableVista benchmark finds foundation models maintain performance across visual styles but degrade sharply on complex table structures and vision-only settings.
FrontierFinance benchmark shows human financial experts outperform state-of-the-art LLMs by achieving higher scores and more client-ready outputs on realistic long-horizon tasks.
BloombergGPT is a 50B parameter LLM trained on a 708B token mixed financial and general dataset that outperforms prior models on financial benchmarks while preserving general LLM performance.
ChunkGroupSHAP clusters semantically related chunks into shared cross-document features for listwise Shapley explanations of embedding-based rankings.
CRAFT is a unified bidirectional counterfactual reasoning framework that improves LLM performance on tabular QA and fact verification tasks over baselines on WikiTQ and TabFact.
OCC-RAG develops task-specialized SLMs (0.6B and 1.7B) via a new synthetic data pipeline for multi-hop reasoning and context faithfulness, claiming to match or exceed 2-6x larger general models on HotpotQA, MuSiQue, TAT-QA, ConFiQA, and MuSiQue-Un.
Using LLM extraction on 681 papers, the authors build a public knowledge graph showing financial NLP moved from LLM adoption to limitation-aware, modular system design between 2022 and 2025.
EvidenceLens is a visual analytics system that decomposes LLM financial answers into atomic claims and visualizes their multimodal evidence alignment, support gaps, and contradictions through a claim-evidence matrix and review-priority ranking.
QuestBench is a student-constructed benchmark of 256 questions on which current deep research AI systems achieve a mean pass rate of 16.85% and a best-case rate of 57.58%.
citing papers explorer
-
Fin-Bias: Comprehensive Evaluation for LLM Decision-Making under human bias in Finance Domain
LLMs copy biased analyst ratings in investment decisions but a new detection method encourages independent reasoning and can improve stock return predictions beyond human levels.
-
TableVista: Benchmarking Multimodal Table Reasoning under Visual and Structural Complexity
TableVista benchmark finds foundation models maintain performance across visual styles but degrade sharply on complex table structures and vision-only settings.
-
FrontierFinance: A Long-Horizon Computer-Use Benchmark of Real-World Financial Tasks
FrontierFinance benchmark shows human financial experts outperform state-of-the-art LLMs by achieving higher scores and more client-ready outputs on realistic long-horizon tasks.
-
BloombergGPT: A Large Language Model for Finance
BloombergGPT is a 50B parameter LLM trained on a 708B token mixed financial and general dataset that outperforms prior models on financial benchmarks while preserving general LLM performance.
-
Listwise Explanation of Embedding-Based Rankings via Semantic Chunk Grouping
ChunkGroupSHAP clusters semantically related chunks into shared cross-document features for listwise Shapley explanations of embedding-based rankings.
-
CRAFT: A Unified Counterfactual Reasoning Framework for Tabular Question Answering and Fact Verification
CRAFT is a unified bidirectional counterfactual reasoning framework that improves LLM performance on tabular QA and fact verification tasks over baselines on WikiTQ and TabFact.
-
OCC-RAG: Optimal Cognitive Core for Faithful Question Answering
OCC-RAG develops task-specialized SLMs (0.6B and 1.7B) via a new synthetic data pipeline for multi-hop reasoning and context faithfulness, claiming to match or exceed 2-6x larger general models on HotpotQA, MuSiQue, TAT-QA, ConFiQA, and MuSiQue-Un.
-
MetaGraph: A Large-Scale Meta-Analysis of GenAI in Financial NLP (2022-2025)
Using LLM extraction on 681 papers, the authors build a public knowledge graph showing financial NLP moved from LLM adoption to limitation-aware, modular system design between 2022 and 2025.
-
EvidenceLens: A Claim-Evidence Matrix for Auditing Financial Question Answering
EvidenceLens is a visual analytics system that decomposes LLM financial answers into atomic claims and visualizes their multimodal evidence alignment, support gaps, and contradictions through a claim-evidence matrix and review-priority ranking.
-
Teaching AI Through Benchmark Construction: QuestBench as a Course-Based Practice for Accountable Knowledge Work
QuestBench is a student-constructed benchmark of 256 questions on which current deep research AI systems achieve a mean pass rate of 16.85% and a best-case rate of 57.58%.