Deeper is not always better in LLMs Alibaba's Qwen team releases Confident Decoding, a training-free inference method that bypasses the final layer's alignment tax by picking the most confident…

On the hardest tier of the Omni-MATH benchmark, gpt-oss-20b scored 1.1%. Same weights, no retraining: it reached 23.5% once its predictions were read from a more confident layer just short of the…

1/8 🧠 Think the deepest layer of an LLM is always the best for output? Think again! Our latest paper by Qwen Team reveals the "Alignment Tax" hiding in your final layers. Post-training can violently…