SimLens for Early Exit in Large Language Models: Eliciting Accurate Latent Predictions with One More Token

arXiv:2507.17618v3 Announce Type: replace 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.

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.

Computation and Language (cs.CL); Performance (cs.PF)

arXiv:2507.17618 [cs.CL]

(or arXiv:2507.17618v3 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2507.17618 Focus to learn more arXiv-issued DOI via DataCite

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

aiqbench analysis

SimLens for Early Exit in Large Language Models: Eliciting Accurate Latent Predictions with One More Token

Related coverage