The paper introduces Adaptive Log‑Space (AL) quantization, a block‑wise representation that adapts the non‑zero range per block and preserves the exact‑zero invariant for non‑negative optimizer states. AL8 and AL16 are combined with signed‑momentum encodings and state‑specific precision choices, rather than a single policy for all states. Experiments on TinyLlama‑1.1B and GPT‑2 show that AL‑based quantization can match or exceed full‑precision performance while dramatically reducing optimizer‑state storage.
arXiv:2608.21952v1 Announce Type: new Abstract: Recent advances in sequence modeling have highlighted Mamba as a state space architecture offering efficient long-range dependency modeling and providi...
arXiv:2608. 08081v1 Announce Type: cross Abstract: Large mixture-of-experts (MoE) language models with 26--120 billion parameters exceed the memory capacity of consumer devices through three simultaneous pressures: resident weight matrices, key-value (KV) cache state that grows linearly with context, and dozens of expert sublayers that must be paged on demand.
arXiv:2605. 24011v2 Announce Type: replace-cross Abstract: Vision-Language-Action (VLA) models exhibit remarkable action generation for embodied intelligence, but their heavy compute make deployment on edge platforms impractical.
arXiv:2607. 02893v1 Announce Type: new Abstract: Low-bit quantization shrinks language models but treats precision as a single global hyper-parameter: every weight uses the same bit-width.
HBQ: Hierarchical Scaling Block Quantization with Hardware‑Efficiency‑Aware Design for Accurate LLM Inference proposes a new block‑quantization scheme that uses large blocks and low‑overhead significand scaling to balance hardware efficiency and accuracy. The authors demonstrate that larger blocks improve efficiency by amortizing dequantization and accumulation costs, while their SIG scaling compensates for the resulting accuracy loss. Experiments on a 28 nm ASIC accelerator show that HBQ achieves up to 4.6× higher area/energy efficiency than state‑of‑the‑art weight‑only quantization, with 1.5–3.0× speedup and 1.6–3.3× system energy reduction over existing BQ methods.
arXiv:2606. 07819v1 Announce Type: new Abstract: Recently, the efficiency of Large Language Models (LLMs) deployment has become a critical concern in practical applications.
arXiv:2607. 28418v1 Announce Type: cross Abstract: Pruning is a promising approach for improving the efficiency of LLMs.
arXiv:2608. 04407v1 Announce Type: cross Abstract: Memory-efficient matrix optimizers such as Sinkhorn gradient descent remove most AdamW optimizer state for dense Transformer matrices, but direct application to Mixture-of-Experts (MoE) training is unreliable.
The paper introduces DAMP, a decay‑aware mixed‑precision quantization scheme for recurrent‑state representations in GDN and KDA language models. By identifying high‑risk channels through quantization‑error energy and decay persistence, DAMP stores these channels at higher precision while compressing the rest to INT8, achieving a 9.9‑bit average precision. Experiments on Qwen3.6‑35B and Kimi‑Linear‑48B show a 69.1% reduction in recurrent‑state storage, up to 2.01× faster state‑update kernels, and up to 10.9% lower full‑model TPOT while preserving accuracy close to the FP32 baseline.
The paper presents a 4‑bit quantization recipe, Minima: NVFP4 W4A4, that fully quantizes all linear layers—including the Gated DeltaNet (GDN) recurrent blocks—of the 27‑billion‑parameter Qwen3.8 LLM. Across a suite of benchmarks (perplexity, MMLU‑Pro, GSM8K, AIME'25, GPQA‑Diamond, LiveCodeBench, and RULER retrieval), the quantized model matches BF16 performance within seed noise while being 17.5 GiB in size and 14–19 % faster at prefill. The authors attribute this success to four mechanisms: block‑scaling of residuals, robust gate projections, the delta‑rule recurrence’s noise‑plateau behavior, and the per‑token quantization cost’s dilution over long contexts.
arXiv:2606. 00079v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) large language models reduce per-token computation through sparse expert activation, but their deployment remains memory-intensive because all expert weights must be kept resident in memory.
arXiv:2607. 08241v1 Announce Type: cross Abstract: Deploying classifier-free guidance (CFG) diffusion models under real-world compute budgets requires quantization, yet existing post-training quantization (PTQ) methods treat CFG models as single-branch networks, ignoring the paired conditional/unconditional structure that CFG inference fundamentally relies on.