MMESGBench
First multimodal benchmark for understanding and complex reasoning over real-world ESG reports; 933 expert-validated QA pairs across 45 documents with text/table/chart evidence and single-page, cross-page, and unanswerable questions (ACM MM 2025).
- from
- AI in Finance
- added
- 2026-10-10
- likes
- 0
similar
-
ESGenius github.com
The first benchmark for evaluating LLMs on Environmental, Social, and Governance (ESG) and sustainability knowledge; 1,136 expert-validated multiple-choice questions plus a curated source corpus, evaluating 50 LLMs under zero-shot and RAG (EMNLP 2025).
-
BDE Score github.com
AI-powered multi-market stock analysis with transparent BDE scoring across 73 stocks (US/HK/A-share). EU AI Act Art.50 compliant. MIT license.
-
GLM github.com
A series of open-source MoE language models by Z.ai for agentic tasks. #opensource
-
ai-evaluation github.com
Open-source LLM and agent evaluation framework with 50+ metrics, LLM-as-Judge augmentation, and guardrail scanners (jailbreak, PII, prompt-injection). Useful for scoring RAG outputs, agent trajectories, and function-calling behavior in data-science workflows.
-
MLEM github.com
-
ESM3 github.com
Multimodal protein language model that jointly reasons over sequence, structure, and function for generative protein design and engineering.
AI in Finance › LLMs: “First multimodal benchmark for understanding and complex reasoning over real-world ESG reports; 933 expert-validated QA pairs across 45 documents with text/table/chart evidence and single-page, cross-page, and unanswerable questions (ACM MM 2025).”