LLM Digest

Story

arxiv_cs_cl · Nov 10, 2026 · paper

Source brief

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

arxiv.orgNov 10, 2026
original source linked

In brief

Local language models can save memory by storing weights in fewer bits, and they can reduce sensitivity to the order of answer options by averaging a few reorderings. We ask whether the first makes the second worse, and which is the cheaper way to spend a fixed budget. On Qwen3-0

Read the original at arxiv.org →Open in live feed Read that day’s brief

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

Earlier in this thread 4 items

Show HN: LLM weight editing tool made by a 15-year-old

Stop running everything through LLMs: How I fixed my research agent

Where Should a KV-Cache Repair Budget Go? Distractor Structure and Contextual Compatibility in Non-Prefix Reuse

Anthropic launched tools to show companies what their AI spending is buying