Question Answering on AdversarialQA (test)
73.2Average AccuracyProCL
Evaluation Results
| Method | Links | |
|---|---|---|
| ProCLLLM=Llama-3.1-8B2026.05 | 73.2 | |
| DEALLLM=Qwen3-8B2026.05 | 72.9 | |
| ProCLLLM=Qwen3-8B2026.05 | 72.8 | |
| ProCLLLM=Qwen3-4B2026.05 | 72.5 | |
| O-LoRALLM=Llama-3.1-8B2026.05 | 72.4 | |
| DEALLLM=Llama-3.1-8B2026.05 | 72.1 | |
| ProCLLLM=Llama-3.2-3B2026.05 | 69.8 | |
| ProCLLLM=Average2026.05 | 69.1 | |
| DEALLLM=Qwen3-4B2026.05 | 68.2 | |
| EWCLLM=Llama-3.2-3B2026.05 | 68.1 | |
| O-LoRALLM=Llama-3.2-3B2026.05 | 68.1 | |
| Seq-LoRALLM=Llama-3.2-3B2026.05 | 67.7 | |
| ReplayLLM=Llama-3.2-3B2026.05 | 67.6 | |
| DEALLLM=Llama-3.2-3B2026.05 | 66.4 | |
| DEALLLM=Average2026.05 | 66.2 | |
| EWCLLM=Llama-3.1-8B2026.05 | 65.3 | |
| Seq-LoRALLM=Llama-3.1-8B2026.05 | 64.5 | |
| ProCLLLM=Flan-T5-Large2026.05 | 63.4 | |
| ReplayLLM=Llama-3.1-8B2026.05 | 62.8 | |
| ProCLLLM=Flan-T5-Base2026.05 | 62.8 | |
| ReplayLLM=Flan-T5-Large2026.05 | 62.8 | |
| EWCLLM=Flan-T5-Large2026.05 | 62.6 | |
| Seq-LoRALLM=Flan-T5-Large2026.05 | 62 | |
| O-LoRALLM=Average2026.05 | 61.9 | |
| EWCLLM=Average2026.05 | 61.7 | |
| O-LoRALLM=Flan-T5-Large2026.05 | 61.6 | |
| DEALLLM=Flan-T5-Base2026.05 | 61.4 | |
| Seq-LoRALLM=Flan-T5-Base2026.05 | 61.3 | |
| Seq-LoRALLM=Average2026.05 | 61.2 | |
| ReplayLLM=Average2026.05 | 61.2 | |
| EWCLLM=Flan-T5-Base2026.05 | 60.9 | |
| ReplayLLM=Flan-T5-Base2026.05 | 60.6 | |
| Seq-LoRALLM=Qwen3-8B2026.05 | 59.1 | |
| ReplayLLM=Qwen3-8B2026.05 | 58.8 | |
| EWCLLM=Qwen3-8B2026.05 | 58.7 | |
| O-LoRALLM=Qwen3-8B2026.05 | 58.6 | |
| O-LoRALLM=Qwen3-4B2026.05 | 56.6 | |
| DEALLLM=Flan-T5-Large2026.05 | 55.9 | |
| ReplayLLM=Qwen3-4B2026.05 | 54.8 | |
| EWCLLM=Qwen3-4B2026.05 | 54.5 | |
| O-LoRALLM=Flan-T5-Base2026.05 | 54.2 | |
| Seq-LoRALLM=Qwen3-4B2026.05 | 52.5 |