SPIN Unprocessed September 3, 2026 ai_technology research
Post-Training Ternarization of Qwen3-4B Capability, Effective Bit Budget, Storage Compression, and Deployment
View original on arxiv.orgOverview
arXiv:2609.01962v1 Announce Type: new Abstract: Ultra-low-bit language models can reduce storage and memory bandwidth, but a nominal "1.58-bit" label does not fully describe the stored representation, retained capability, or runtime behavior. We study an end-to-end post-training conversion of Qwen, an instruction-tuned 4B-parameter model, using KOTMS rotation, E2M-ATQ ternarization, and GPTQ-style error compensation from TWLA. The experiment is weight-only: activations remain at 16-bit precision
SpinGraph analysis pending — check back after processing.
Ask AI about this story
Opens with the SpinGraph .md URL and structured context — one click, prompt included.
More from arXiv Artificial Intelligence
View all →- MineTRACE: An Evidence-Grounded Interactive Reasoning System for Mineral Prospectivity
- DocHop: Benchmarking Out-of-domain Multi-hop Reasoning in Information-Dense Documents
- Monitoring Web Agents Without Internal Signals: Observable Trajectories and Key-Step Supervision
- HeadWiseKV: Budgeted Per-Head Cache Residency for Hybrid Long-Context Language Models
- ClaimReceipt: Verifying Evidence Sufficiency and Coverage in Agent Evaluations
- When Agents Implement Systems: A Case Study in Defects, Detection, and Evaluation Rigor
Markdown (.md) · JSON-LD schema (.json) · Machine-readable for AI & GEO