Table Question Answering on WTQ (Evaluation Set)
47Alignment ScoreLLAMA-3.1-8B-INST
Evaluation Results
| Method | Links | |
|---|---|---|
| LLAMA-3.1-8B-INSTInput Protocol=Vanilla2026.05 | 47 | |
| QWEN2.5-14B-INSTInput Protocol=Vanilla2026.05 | 47 | |
| TABLELLM-8BInput Protocol=Vanilla2026.05 | 33 | |
| QWEN2.5-14B-INSTInput Protocol=Rand Perm2026.05 | 33 | |
| LLAMA-3.1-8B-INSTInput Protocol=Rand Perm2026.05 | 31 | |
| QWEN2.5-7B-INSTInput Protocol=Vanilla2026.05 | 29 | |
| LLAMA-3.1-8BInput Protocol=Vanilla2026.05 | 26 | |
| QWEN2.5-14B-INSTInput Protocol=ATP Attack2026.05 | 26 | |
| TABLELLM-8BInput Protocol=Rand Perm2026.05 | 23 | |
| DS-R1-DIST-LLAMA-8BInput Protocol=Vanilla2026.05 | 23 | |
| LLAMA-3.1-8B-INSTInput Protocol=ATP Attack2026.05 | 22 | |
| QWEN2.5-7B-INSTInput Protocol=Rand Perm2026.05 | 21 | |
| CODELLAMA-7B-INSTInput Protocol=Vanilla2026.05 | 18 | |
| LLAMA-3.1-8BInput Protocol=Rand Perm2026.05 | 17 | |
| TABLELLM-8BInput Protocol=ATP Attack2026.05 | 16 | |
| CODELLAMA-7B-INSTInput Protocol=Rand Perm2026.05 | 16 | |
| DS-R1-DIST-LLAMA-8BInput Protocol=Rand Perm2026.05 | 15 | |
| DS-R1-DIST-QWEN-7BInput Protocol=Vanilla2026.05 | 14 | |
| LLAMA-3.1-8BInput Protocol=ATP Attack2026.05 | 13 | |
| QWEN2.5-7B-INSTInput Protocol=ATP Attack2026.05 | 12 | |
| DS-R1-DIST-QWEN-7BInput Protocol=Rand Perm2026.05 | 12 | |
| CODELLAMA-7B-INSTInput Protocol=ATP Attack2026.05 | 9 | |
| DS-R1-DIST-LLAMA-8BInput Protocol=ATP Attack2026.05 | 9 | |
| DS-R1-DIST-QWEN-7BInput Protocol=ATP Attack2026.05 | 6 |