| { |
| "feat_dim": 601, |
| "props": [ |
| "Egc" |
| ], |
| "polybert": "HAYDERphd/polyBERT", |
| "results": { |
| "baselines": { |
| "single_step": { |
| "greedy_oracle": { |
| "n": 120, |
| "success": 0.75, |
| "regret": 0.251017332402737 |
| }, |
| "greedy_verifier": { |
| "n": 120, |
| "success": 0.30833333333333335, |
| "regret": 0.4944557918330346 |
| }, |
| "random": { |
| "n": 120, |
| "success": 0.075, |
| "regret": 1.6162096724265453 |
| } |
| }, |
| "multi_step": { |
| "greedy_oracle": { |
| "n": 44, |
| "success": 0.22727272727272727, |
| "regret": 0.8539263608315685 |
| }, |
| "greedy_verifier": { |
| "n": 44, |
| "success": 0.022727272727272728, |
| "regret": 1.1261625219993852 |
| }, |
| "random": { |
| "n": 44, |
| "success": 0.022727272727272728, |
| "regret": 2.0974685791787597 |
| } |
| } |
| }, |
| "sft": { |
| "single_step": { |
| "n": 120, |
| "success": 0.8833333333333333, |
| "regret": 0.1476266867135324 |
| }, |
| "multi_step": { |
| "n": 44, |
| "success": 0.8181818181818182, |
| "regret": 0.3510164215163693 |
| } |
| }, |
| "rl": { |
| "single_step": { |
| "n": 120, |
| "success": 0.8583333333333333, |
| "regret": 0.1600647209968158 |
| }, |
| "multi_step": { |
| "n": 44, |
| "success": 0.7727272727272727, |
| "regret": 0.4018683559774568 |
| } |
| }, |
| "props": [ |
| "Egc" |
| ], |
| "verifier": "finetuned_polyBERT", |
| "val_rmse": { |
| "Egc": 0.5077038207039136 |
| }, |
| "n_train_req": 375, |
| "n_steps": 1262 |
| } |
| } |