AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

LAS: Loss-less ANN-SNN Conversion for Fully Spike-Driven Large Language Models

Long Chen, Xiaotian Song, Yanan Sun

AAAI 2026 · Large language models › Language, safety and interpretability · approach: LLM unlearning and benchmarks · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace OAT neuron with a single multi-threshold neuron (ablation)LAS · RealWorldQA, Qwen2-VL-7Baccuracy: -3.36 drop without OATImproved
add spike-equivalent self-attention (ablation)LAS · RealWorldQA, Qwen2-VL-7Baccuracy: 15.77 -> 66.79Improved
use 16 time steps instead of 11 (ablation)LAS · GLUE QNLI, BERT-baseaccuracy: 86.14 -> 90.92Improved
use 13 instead of 16 time steps (ablation)LAS · GLUE QNLI, BERT-baseaccuracy: 90.92 -> 90.91No clear effect
increase MT neuron threshold levels H from 1 to 10 (ablation)LAS · BERT STS-Benergy ratio: 1.03 -> 0.41Improved
increase H (ablation)LAS · BERT STS-Baccuracy: 88.73 to 88.80No clear effect
convert OPT-66B to SNN with LASOPT-66B · WSC (zero-shot)accuracy: 82.78 -> 85.71Improved
convert OPT-66B to SNN with LASOPT-66B · Winogrande (zero-shot)accuracy: 69.98 -> 68.27Hurt
convert LLaVA1.5-7B with LASLLaVA1.5-7B · HallusionBenchscore: 81.36 -> 67.93Hurt
convert Qwen2-VL-7B with LASQwen2-VL-7B · BLINK, MMStaraccuracy: 51.55 -> 52.18 (BLINK)Improved
use LAS instead of FAS conversion on OPT-7BFAS · PIQA (zero-shot)accuracy: 73.23 -> 76.22Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy