Classification 벤치마크
Classification on SST-2
Test Accuracy
- 2024-05-24 — OPT-1.3B: Test Accuracy 90.78
| Rank | Model | Test Accuracy | Paper | Code | Year |
|---|---|---|---|---|---|
| 1 | OPT-1.3B | 90.78% | Achieving Dimension-Free Communication in Federated Learning via Zeroth-Order Optimization | ZidongLiu/DeComFL | 2024 |
| 2 | Greedy Is a Strong Default: Agents as It 자동 추출 | 89.5 | Greedy Is a Strong Default: Agents as Iterative Optimizers | 2026 | |
| 3 | CLAQS 자동 추출 | 87.08 | CLAQS: Compact Learnable All-Quantum Token Mixer with Shared-ansatz for Text Classification | 2025 | |
| 4 | Kathleen 자동 추출 | 85.8 | Kathleen: Oscillator-Based Byte-Level Text Classification Without Tokenization or Attention | 2026 | |
| 5 | OPT-125M | 85.08% | Achieving Dimension-Free Communication in Federated Learning via Zeroth-Order Optimization | ZidongLiu/DeComFL | 2024 |
| 6 | QSLM 자동 추출 | 84.4 | QSLM: A Performance- and Memory-aware Quantization Framework with Tiered Search Strategy for Spike-driven Language Models | 2026 |