Update README.md

Update README.md

Update README.md

fix: update Short comparison table with verified single-model scores

fix: update Long comparison table with 7 verified single-model scores

fix: update Long comparison table to verified single-model scores only

fix: correct per-task nDCG@10 scores (Short from MRE-T1, Long from MRE-T1_long)

fix: restore original model card, only change base_model to Qwen3-4B-Instruct-2507

fix: restore Long comparison table

fix: remove unverified Long competitor scores, keep only MRE-T1

fix: replace Long comparison table with verified single-model scores

fix: correct base_model to Qwen3-4B-Instruct-2507

fix: correct per-task scores to match no-instruction variant (avg 39.6/35.1)

fix: use confirmed model card with company intro and full comparison tables

fix: update citation author to Forward AI

Add model card for MRE-T1

initial commit