Advancing ESG Intelligence: An Expert-level Agent and Comprehensive Benchmark for Sustainable Finance
cs.AI
Submitted: 2026-01-13
Updated: 2026-11-14
Code: https://github.com/ElaineZhao92/ESGAgent-and-Benchmark
Project page: https://www.boeing.com/innovation/sustainability/santa-susana-field-laboratory/a-bright-future-today
Terminology
Sources
- Denoising ESG: quantifying data uncertainty from missing data with Machine Learning and prediction intervals
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- LightRAG: Simple and Fast Retrieval-Augmented Generation
- ESGenius: Benchmarking LLMs on Environmental, Social, and Governance (ESG) and Sustainability Knowledge
- FinSight: Towards Real-World Financial Deep Research
- DeepSeek-V3 Technical Report
- AgentBench: Evaluating LLMs as Agents
- FaithEval: Can Your Language Model Stay Faithful to Context, Even If "The Moon is Made of Marshmallows"
- Towards Robust ESG Analysis Against Greenwashing Risks: Aspect-Action Analysis with Cross-Category Generalization
- ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs
- Correctness is not Faithfulness in RAG Attributions
- ClimateBert: A Pretrained Language Model for Climate-Related Text
- BloombergGPT: A Large Language Model for Finance
- CiteCheck: Towards Accurate Citation Faithfulness Detection
- WebArena: A Realistic Web Environment for Building Autonomous Agents
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents