{"ID":2891415,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.17618","arxiv_id":"2507.17618","title":"SimLens for Early Exit in Large Language Models: Eliciting Accurate Latent Predictions with One More Token","abstract":"Intermediate-layer predictions in large language models (LLMs) are informative but hard to decode accurately, especially at early layers. Existing lens-style methods typically rely on direct linear readout, which is simple but often drifts away from the model's eventual prediction. We proposeSimLens, a simple training-free decoder for single-token decision tasks that keeps only the start token and a candidate answer token ([s] and [a]) and performs one lightweight continuation through the remaining upper layers. This surprisingly small modification recovers much more accurate latent predictions than direct linear decoding. We further introduce Linear SimLens, a lightweight linear approximation for entropy-based confidence estimation, and combine the two in SimExit, a hybrid early-exit mechanism. On ARC, BoolQ, and HeadQA with LLaMA-7B and Vicuna-7B, SimLens improves Iso-Compute accuracy in all six settings, with an average gain of +0.43 even when fair compute includes the extra two-token post-forward overhead. SimExit yields an average 1.15$\\times$ speedup at the best-accuracy operating points and 1.40$\\times$ when allowing up to a 1 percentage-point accuracy drop. Ablations show that [s] and [a] play distinct roles as global condition and semantic anchor, respectively.","short_abstract":"Intermediate-layer predictions in large language models (LLMs) are informative but hard to decode accurately, especially at early layers. Existing lens-style methods typically rely on direct linear readout, which is simple but often drifts away from the model's eventual prediction. We proposeSimLens, a simple training-...","url_abs":"https://arxiv.org/abs/2507.17618","url_pdf":"https://arxiv.org/pdf/2507.17618v2","authors":"[\"Ming Ma\",\"Bowen Zheng\",\"Zhongqiao Lin\",\"Tianming Yang\"]","published":"2025-07-23T15:49:03Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.PF\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}