arXiv Computation and Language August 27, 2026 By Ming Ma, Bowen Zheng, Zhongqiao Lin, Tianming Yang

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

Read the original on arXiv Computation and Language →

SimLens is a training‑free decoder that improves early‑layer predictions in large language models by keeping only the start token and a candidate answer token and performing a lightweight continuation through the remaining layers. It outperforms direct linear readouts, yielding higher accuracy on tasks such as ARC, BoolQ, and HeadQA with LLaMA‑7B and Vicuna‑7B. The method is extended to Linear SimLens for confidence estimation and combined into SimExit, a hybrid early‑exit mechanism that achieves significant speedups while maintaining accuracy.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

llms

arXiv Machine Learning

Sep 30

SEED: Self-Speculative Decoding via Implicit Encoder-Decoder

arXiv:2609.36590v1 Announce Type: cross Abstract: Self-speculative decoding accelerates large language model (LLM) inference by drafting tokens from the target model itself, but faces a sharp tradeof...

By Hankun Lin, Patrick Pynadath, Ruqi Zhang

llms fine-tuning benchmarks

arXiv Computation and Language

Sep 4

Breaking the Likelihood Trap: Variance-Calibrated Modulation for Large Language Model Decoding

The paper introduces Variance‑Calibrated Modulation (VCM), a training‑free pre‑decoding technique that reshapes language model probability distributions before truncation. VCM uses two dynamic mechanisms: a Contextual Searchlight via PMI to suppress stopwords and highlight context‑relevant tokens, and an Adaptive Self‑Debiasing that applies scale‑invariant penalization based on real‑time logit standard deviation. Experiments on open‑ended generation, factual QA, and mathematical reasoning show that VCM consistently reduces the likelihood trap, improving diversity, coherence, and reasoning accuracy with minimal computational cost.

By Yuanhao Ding, Meimingwei Li, Esteban Garces Arias, Matthias A{\ss}enmacher, Christian Heumann, Chongsheng Zhang

llms

arXiv Machine Learning

Jul 13

Accelerating Large Language Model Inference with Self-Supervised Early Exits

arXiv:2407. 21082v3 Announce Type: replace-cross Abstract: This paper presents a modular approach to accelerate inference in large language models (LLMs) by adding early exit heads at intermediate transformer layers.

By Florian Valade

llms benchmarks

arXiv Computation and Language

Sep 1

Manac\'a-1B: An Open, Reproducible Brazilian-Portuguese Language Model and a Tokenizer-Aware, Paired Evaluation

Manacá-1B is a 1.72‑billion‑parameter, open decoder‑only language model trained from scratch for Brazilian Portuguese, released with a fully containerized, reproducible training pipeline and complete logs. The authors evaluate it against nine open baselines on four Portuguese benchmarks, reporting standard errors and paired significance tests, and find that Manacá-1B outperforms smaller models on LAMBADA‑PT while remaining competitive on commonsense completion. They also uncover a tokenizer‑related evaluation pitfall that can drastically lower accuracy and provide a simple fix, releasing all code, logs, and corrected tokenizer for full reproducibility.

By Bruno Leonardo Santos Menezes, Carlos Leonardo Souza Cardoso, Fabio Andre Machado Porto

llms nlp benchmarks

arXiv AI

Jun 16

Follow the Latent Roadmap: Navigating Revocable Decoding for Diffusion LLMs with Anchor Tokens

arXiv:2606. 16847v1 Announce Type: cross Abstract: Diffusion Large Language Models (dLLMs) offer a promising avenue for parallel generation but face a trade-off between decoding speed and quality.

By Yizhen Yao, Qinglin Zhu, Runcong Zhao, Xiangxiang Dai, Yanzheng Xiang, Yulan He, Lin Gui

llms rag diffusion efficiency benchmarks

arXiv AI

Jun 2

SimSD: Simple Speculative Decoding in Diffusion Language Models

arXiv:2606. 02544v1 Announce Type: cross Abstract: Diffusion large language models (dLLMs) have recently emerged as a promising alternative to autoregressive (AR) LLMs, offering faster inference through parallel or blockwise decoding.

By Junxia Cui, Haotian Ye, Runchu Tian, Hongcan Guo, Jinya Jiang, Haoru Li, Chaojie Ren, Yiming Huang, Kaijie Zhu, Zhongkai Yu, Kun Zhou, Jingbo Shang

llms diffusion efficiency benchmarks