Seat-swap comparisons
4 of 32 matched pair(s), paired on
instance_id, seed, arm and selected by largest-effect. 32 episode(s) left, 32 right, 32 shared key(s). The primary/USW/worst-off columns are DELTAS (right minus left), not levels.| page | model | arm | visibility | instance | difficulty | parameter tags | seed | outcome | primary | dist NBS | USW | worst-off | fabricated | hazards | score Δ | total regret |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| scorable_negotiation-L0-4abed256fd seed 1 · moves_chat · no seat swap | Qwen3-8B vs Qwen3-8B | moves_chat | PRIVATE | scorable_negotiation-L0-4abed256fd | — | — | 1.000 | deal | 1.000 | — | 57.0 | 8.000 | 0% | 76.92% silent budget 32768 | 1.000 | — |
| scorable_negotiation-L0-b7ecc0292a seed 0 · moves_chat · no seat swap | Qwen3-8B vs Qwen3-8B | moves_chat | PRIVATE | scorable_negotiation-L0-b7ecc0292a | — | — | 0.000 | deal | 1.000 | — | 45.2 | 5.250 | 0% | 71.43% silent 1 no-ballot budget 32768 | 1.000 | — |
| scorable_negotiation-L0-2b42ce4a5c seed 0 · moves_chat · no seat swap | Qwen3-8B vs Qwen3-8B | moves_chat | PRIVATE | scorable_negotiation-L0-2b42ce4a5c | — | — | 0.000 | deal | 1.000 | — | 42.0 | 1.000 | 0% | 78.57% silent 1 no-ballot budget 32768 | 1.000 | — |
| scorable_negotiation-L0-2d5d5b106e seed 1 · moves_chat · no seat swap | Qwen3-8B vs Qwen3-8B | moves_chat | PRIVATE | scorable_negotiation-L0-2d5d5b106e | — | — | 1.000 | deal | 1.000 | — | 33.0 | 0.000 | 0% | 69.23% silent budget 32768 | 1.000 | — |
Click a column header to sort; / focuses the filter, Enter opens the first row that survives it.