The study examines how user context—such as memory, profiles, and role prompts—affects Large Language Models’ (LLMs) financial analysis. Using 3,575 SEC filings and twelve LLMs, the authors distinguish between evidence selection and interpretation, finding that most context spillover arises from differing interpretations under various roles rather than from retrieving different evidence. They evaluate two mitigation strategies—expressing investor mindset as a user profile instead of an assistant role, and separating evidence-based from personalized outputs—both of which reduce but do not eliminate spillover, with effectiveness varying across models.
By Ahmed Asaad, Amr Mohamed, Yang Zhang, Omneya Abdelsalam
arXiv:2602. 07294v4 Announce Type: replace-cross Abstract: With the increasing deployment of Large Language Models (LLMs) in the finance domain, LLMs are increasingly expected to parse complex regulatory disclosures.
By Yidong Jiang, Junrong Chen, Eftychia Makri, Jialin Chen, Peiwen Li, Ali Maatouk, Leandros Tassiulas, Eliot Brenner, Bing Xiang, Rex Ying
arXiv:2606. 29251v1 Announce Type: new Abstract: Financial decision-makers face more information than they can directly inspect, making context compression necessary.
By Hoyoung Lee, Suhwan Park, Seunghan Lee, Jun Seo, Jaehoon Lee, Sungdong Yoo, Minjae Kim, CheolWon Na, Zhangyang Wang, Zach Golkhou, Minkyu Kim, Sotirios Sabanis, Alejandro Lopez-Lira, Dhagash Mehta, Soonyoung Lee, Chanyeol Choi, Wonbin Ahn, Yongjae Lee
The paper investigates how user context—such as memory, profiles, and role prompts—affects large language models’ financial analysis. By testing 3,575 SEC filings across twelve LLMs, the study distinguishes between evidence selection and interpretation, finding that interpretation under different roles drives most user-context spillover. Two mitigation strategies—using a user profile instead of an assistant role and separating evidence-based from personalized outputs—reduce but do not eliminate this spillover, with effectiveness varying by model.
arXiv:2606. 23032v2 Announce Type: replace Abstract: Finance Agent v2 (by Vals AI) has emerged as the reference benchmark for evaluating both Anthropic Claude and OpenAI ChatGPT frontier language models on financial tasks.
By Mostapha Benhenda
Finance Agent v2 (by Vals AI) has emerged as the reference benchmark for evaluating both Anthropic Claude and OpenAI ChatGPT frontier language models on financial tasks. However, it narrowly deals with periodic reporting from publicly traded companies (SEC 10-K and 10-Q filings), and its agentic harness relies on naive, unenriched chunk retrieval.
arXiv:2606. 23032v3 Announce Type: replace Abstract: Finance Agent v2 (by Vals AI) has emerged as the reference benchmark for evaluating both Anthropic Claude and OpenAI ChatGPT frontier language models on financial tasks.
By Mostapha Benhenda
FinRiskAtlas is a Chinese-language benchmark designed to evaluate large language models (LLMs) for financial risk review by focusing on decision‑aligned tasks rather than generic financial knowledge. It contains 9,742 instances across 53 task families, including 42 domain‑knowledge families and 11 downstream review operations defined by explicit evaluation contracts. The extended FinRisk‑Ask framework replays 680 pre‑action states from 104 professional trajectories, withholding future evidence during inference to assess evidence‑state control and request targeting. Results across 33 model configurations show that operation‑level evaluation yields distinct rankings and that knowledge‑based shortlisting can incur significant regret, while frequent use of the Ask branch does not necessarily improve evidence acquisition, highlighting gaps in broad financial capability scores.
By Suyang Zhong, Jingzhe Zhu, Qi Xu, Liyao Sun, Yin Wang, Qingqing Sun, Shuai Chen, Tianyi Zhang
Financial disclosures contain numerical claims, temporal statements, entity references, policy commitments, and risk descriptions that may conflict in qualitatively different ways. Detecting a conflict is only the first step: review workflows may also need to determine its type, since numerical, temporal, referential, factual, and normative inconsistencies require different evidence and downstream checks.
arXiv:2608.22852v1 Announce Type: new
Abstract: Large language models (LLMs) are increasingly used in investment decision-making, yet prior work shows that they exhibit systematic, model-specific inv...
By Sahong Park, Suhwan Park, Hoyoung Lee, Gakyung Kwon, Wonbin Ahn, Jaewon Choi, Alejandro Lopez-Lira, Yoon Kim, Chanyeol Choi, Hyeongwoo Kong, Yongjae Lee
arXiv:2607. 26368v1 Announce Type: cross Abstract: Financial disclosures contain numerical claims, temporal statements, entity references, policy commitments, and risk descriptions that may conflict in qualitatively different ways.
By Aman Kumar, Lasitha Vidyaratne, Dipanjan D Ghosh, Arnab Chakrabarti, Ahmed K Farahat
FinSkillBench is an evaluation suite that tests whether language model agents can use financial domain skills to solve investment management tasks across portfolio construction, risk management, and fundamental analysis. The benchmark contains 12 subtasks with 2,603 episodes, each providing point‑in‑time inputs, hidden ground truth, and a verifier. Experiments show that curated skill packages improve performance significantly, while self‑generated skills offer little benefit, indicating that reliable procedural skills are crucial for effective AI agents in this domain.
By Jermyn Zhen Yong Bek, Zhuang Qiang Bok, Zhongtian Sun