AI Paper Atlas

Research map / Architectures and efficiency / Quantization

Equilibrium Language Models

Yikun Jiang, Huanyu Wang, Tianhong Ding, Wenhu Zhang, Yiming Wu, Hanbin Zhao, John C.S. Lui

ICLR 2026 · Architectures and efficiency › Quantization · approach: KV cache quantization

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
add One-Step KV cache (vs Multi-Step) (ablation)ELM Qwen2.5-1.5B-Instruct M=8 · GSM8Kaccuracy: 71.4 -> 71.8No clear effect
add One-Step KV cache (vs Multi-Step) (ablation)ELM Qwen2.5-1.5B-Instruct M=14 · KV cache memory (4K context)KV (MB): 114.8 -> 61.6Improved
add One-Step KV cache (vs Multi-Step) (ablation)ELM Qwen2.5-1.5B-Instruct M=8 · KV cache memory (4K context)KV (MB): 114.8 -> 86.0Improved
use GPPO to pick start layer instead of cosine similarity (ablation)ELM Qwen2.5-1.5B-Instruct · MBPPpass@10: 65.4 -> 70.6Improved
use GPPO to pick start layer instead of perplexity (ablation)ELM Qwen2.5-1.5B-Instruct · MBPPpass@10: 66.6 -> 70.6Improved
use transformer layer instead of FFN as fixed-point module (ablation)ELM Qwen2.5-1.5B-Instruct · MMLUaccuracy: 51.6 -> 54.0Improved
initialize Wh as identity instead of random (ablation)ELM Qwen2.5-1.5B-Instruct · MBPPpass@10: 69.8 -> 70.6Improved
train with SJFB (ablation)ELM Qwen2.5-1.5B-Instruct · training cost (math)GPU-days: 7.1 -> 6.1Improved
prune one continuous group instead of two disjoint groups (M=14) (ablation)ELM Qwen2.5-1.5B-Instruct · MBPPpass@10: 58.4 -> 65.4Improved
prune one continuous group instead of two disjoint groups (M=14) (ablation)ELM Qwen2.5-1.5B-Instruct · HumanEvalpass@10: 56.7 -> 58.5Improved
use Anderson acceleration instead of simple iteration (M=14, delta 0.5)ELM Qwen2.5-1.5B-Instruct · GSM8K iterationsiterations: 12.25 -> 7.52Improved
use Broyden solver instead of simple iteration (M=8)ELM Qwen2.5-1.5B-Instruct · GSM8Kaccuracy: 71.8 -> 69.8Hurt

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