AI Paper Atlas

Research map / Time series / Foundation models for time series

Learning the ESG Geometry with Domain Aware Language Models

Kunal Pradeep Pimparkhede, Chirayu Chaurasia, Jatin Roy, Mahesh Mohan M R

ICML 2026 · Time series › Foundation models for time series · approach: Time series foundation models

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add FACET tokens and geometry loss (ablation)DA-LLM · ESG risk forecastMSE: 8.866 -> 5.227Improved
add geometry loss on top of FACET tokens (ablation)DA-LLM · ESG risk forecastMSE: 9.321 -> 5.227Improved
add news text (ablation)DA-LLM · ESG risk forecastMSE: 27.643 -> 5.227Improved
add news sentiment (ablation)DA-LLM · ESG risk forecastMSE: 38.56 -> 5.227Improved
add GRI news filtering (ablation)DA-LLM · ESG risk forecastMSE: 17.289 -> 5.227Improved
add LoRA fine-tuning on value-aware tokens (ablation)DA-LLM · ESG risk forecastMSE: 86.97 -> 5.227Improved
add value-aware domain tokens (VADT) (ablation)DA-LLM · ESG risk forecastMSE: 128.97 -> 5.227Improved
set geometry loss weight beta to 0.4 instead of 0 (ablation)DA-LLM · ESG risk forecastMSE: 8.866 -> 5.227Improved
raise geometry loss weight beta from 0.4 to 0.8 (ablation)DA-LLM · ESG risk forecastMSE: 5.227 -> 29.741Hurt
add SAP before NSGA-II optimization (ablation)NSGA-II portfolio rebalancing · Portfolio ESG riskESG risk: 46.5 -> 37.1Improved
select by Euclidean distance instead of Nash criterion for NSGA2 (ablation)NSGA2 with SAP · Portfolio ESG riskESG risk: 39.67 -> 37.1Improved
use DA-LLM instead of iTransformerDA-LLM · ESG risk forecastMSE: 7.334 -> 5.227Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy