arXiv AI

Precision on Demand: Escalating Weight Bits at Uncertain Decoding Steps in a Small Local Model

The paper proposes a dynamic weight‑precision strategy for local decoding, storing 8‑bit weights but using only the top 4 bits during most steps and escalating to full precision when the low‑precision output distribution is uncertain (high entropy or small top‑two margin). Experiments on Qwen3‑0.6B with the GSM8K benchmark show that a 15% escalation budget captures 78% of low‑precision argmax flips, outperforming a random gate, and that an entropy gate at 16% escalation achieves 62.7% accuracy with only 0.87 of the weight‑byte cost of using all‑high precision. The study demonstrates that selective precision escalation can reduce computational cost while maintaining competitive performance.

arXiv AI
6d ago

Bits or permutations? Spending a fixed inference budget on weight precision or on option-order averaging in small local language models

The study investigates how to best allocate a fixed inference budget between reducing weight precision (quantization) and averaging over different answer‑option orders in small local language models. Experiments on Qwen3‑0.6B and Qwen3‑1.7B show that 6‑bit quantization retains or even slightly improves accuracy compared to fp16, while 4‑bit quantization introduces significant answer‑slot skew. Averaging four cyclic option orders can recover a large accuracy gain at 4 bits, and cost‑matched analysis indicates that using 6‑bit weights plus up to four order averages is the most efficient strategy when the budget allows at least 6 bits per weight.

By Mohammed Nayeem Khan, Logesh L, Aliyas Ahmad Misri
arXiv AI
3d ago

One Model, Two Precisions: Does Disagreement Between Quantized Copies Tell You When a Small Model Is Wrong?

The paper investigates whether disagreement between two quantized versions of the same language model can serve as an error signal for tasks such as abstention and escalation. Using Qwen3-0.6B and Qwen3-1.7B, the authors compare an 8‑bit copy with lower‑precision copies (3–6 bits) on multiple-choice datasets, measuring error detection via AUROC, risk‑coverage, and ECE. Results show modest gains over entropy alone, with additional forward passes offering limited improvement, and the study provides a deployment recipe and discusses scenarios where the method is ineffective.

By Mohammed Nayeem Khan, Allan Kevin T
arXiv Machine Learning
Oct 28

ScaleRLT: Exact Log-Probabilities and Fast Deferred Rollouts for Reinforcement Learning on Looped Transformers

ScaleRLT introduces a hybrid learner log‑probability method that reconciles differences between dense and deferred rollout computations in looped language models like Ouro. By using a stop‑gradient replay of the rollout schedule, the approach achieves bit‑for‑bit reproducibility across FP32 and bf16 real‑weight cells while maintaining the same gradient accuracy as dense implementations. The technique boosts rollout throughput by up to 1.56× on a single GPU and significantly speeds up multi‑node GRPO updates, leading to substantial accuracy gains on GSM8K in short training runs.

By Dezhen Lu, Zixiang Ni, Runguo Li
arXiv AI
Oct 28

Channel Selection for Mixture-of-Experts Decoding Needs No Calibration Data

The paper demonstrates that Mixture-of-Experts (MoE) models can avoid the need for calibration data when selecting which expert channels to load during decoding. Instead of using a calibrated threshold, a top‑k selection strategy—choosing the k highest‑scoring channels per token—achieves comparable or better performance without calibration. Additionally, the authors propose a reader that uses a ternary sketch and low‑rank mapping to recover skipped channels, achieving 98.1% of full‑model accuracy while reading only 17.8% of expert weights and speeding up inference by up to 2.69× compared to optimized vLLM.

By Geunhyoung Kim, Sung Ju Hwang
Hugging Face Trending Papers
5d ago

The Noise Floor of a Quantization Comparison: Engine, restart, batch and kernel drift, measured against 8-bit and 4-bit effects in two decision models

The paper investigates how various non‑quantization factors—such as engine changes, server restarts, batching differences, and kernel selection—affect the performance of quantized language models compared to their BF16 baselines. Using 2,400 public decisions from Cloudflare’s Clef‑flash (9B) and Clef (27B) models on an NVIDIA GB10, the authors quantify the noise introduced by each factor and isolate the true impact of different quantization schemes (8‑bit, FP8, Q4_K_M, NVFP4). They find that non‑quantization sources contribute up to 0.92% of decision changes, while 4‑bit quantization can alter accuracy by up to 0.95 points, and they provide code, recipes, and detailed per‑record outputs for reproducibility.

arXiv Machine Learning
Oct 12

DuoRate: Two-Rate Per-Expert Precision Allocation for Mixture-of-Experts Serving

The paper introduces DuoRate, a method for allocating precision in Mixture-of-Experts (MoE) models that separates storage and decoding rates. It stores a uniform tier of weights once and then adapts read precision per workload using a Lagrangian rate‑distortion optimization, reducing the store‑read gap. Experiments on Qwen3-30B-A3B and GSM8K show that DuoRate achieves higher accuracy at lower bit‑per‑weight (bpw) rates compared to fixed‑format baselines.

By Mengqi Guo, Zhe Wei, Yuan Yuan, Jiunn Bin Lim, Boyi Pan, Michael Bi Mi