mlsanityData to 14 Oct 2026

The Fourth State: Signed-Zero Ternary for Stable LLM Quantization (and More)

arXiv 2508.05905 ↗First posted 8 Aug 2025cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Quantization is usually regarded as a means to trade quality of performance for reduced compute requirements, i.e., as a suboptimal approximation. However, if examined in terms of a fixed overall resource budget, a very different perspective arises. We introduce Signed-Zero Ternary (SZT), a 2-bit quantization that deterministically provides gradient information with no forward-path penalty. Our analysis provides evidence that it may improve information density compared to non-quantized alternatives.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 6 of its 13 references that are in the map. First posted on arXiv 8 Aug 2025.

Builds on

6 of 13 references.

References · 6

  1. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,079
  2. 2022GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformerscited by 1,919
  3. 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,593
  4. 2020Up or Down? Adaptive Rounding for Post-Training Quantizationcited by 621
  5. 2016DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradientscited by 569
  6. 2016Trained Ternary Quantizationcited by 208

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026CAT-Q: Cost-efficient and Accurate Ternary Quantization for LLMs
  2. 2025iFairy: the First 2-bit Complex LLM with All Parameters in ±1, ± i
  3. 2025ParetoQ: Improving Scaling Laws in Extremely Low-bit LLM Quantization
  4. 2025Tequila: Trapping-free Ternary Quantization for Large Language Models
  5. 2026BCJR-QAT: A Differentiable Relaxation of Trellis-Coded Weight Quantization
  6. 2025SignRoundV2: Toward Closing the Performance Gap in Extremely Low-Bit Post-Training Quantization for LLMs
  7. 2026Sherry: Hardware-Efficient 1.25-Bit Ternary Quantization via Fine-grained Sparsification
  8. 2024QuIP#: Even Better LLM Quantization with Hadamard Incoherence and Lattice Codebooks