ZICQ curates daily AI news and deep dives on models, open-source infrastructure, and developer field reports.
Submit an articleCurrent filter: Tag: #数据透明 Clear filters
Research & Papers· 2026-10-16#Large Language Models#Symbolic Regression#Structural Verification#LLMs & Foundation Models#arXiv
By Mr.Xu0 views0 discussions
L2G-SR is a novel framework for LLM-based symbolic regression that introduces a learned st…
LLMs & Foundation Models· 2026-10-16
Mr.Xu0 views0 discussions
arXiv has introduced a novel linear-time Transformer architecture based on Taylor series e…
Research & Papers· 2026-10-16
Mr.Xu0 views0 discussions
IneqAgent is a novel agentic framework that revolutionizes formal inequality proving by in…
LLMs & Foundation Models· 2026-10-16
Mr.Xu0 views0 discussions
This research provides an in-depth analysis of the long-horizon reliability of the Qwen3-8…
LLMs & Foundation Models· 2026-10-16
Mr.Xu0 views0 discussions
arXiv introduces the Counterfactual Neighborhood Risk Fields (CNRF) framework, a novel app…
Research & Papers· 2026-10-16
Mr.Xu0 views0 discussions
The Verus team introduces a novel phase-separated completeness auditing approach that miti…
Agentic· 2026-10-16
Mr.Xu0 views0 discussions
AlphaDiana is a measurement system designed to evaluate the performance of agents in verif…
Research & Papers· 2026-10-16
Mr.Xu0 views0 discussions
VISTA is a training-free method that optimizes sparse guidance schedules using dynamic pro…
Research & Papers· 2026-10-16
Mr.Xu0 views0 discussions
As AI model capabilities advance rapidly, benchmarks are becoming saturated, diminishing t…
LLMs & Foundation Models· 2026-10-16
Mr.Xu0 views0 discussions
CoTS (Competitive Token Selection) is a novel training-free framework that dynamically bal…
LLMs & Foundation Models· 2026-10-16
Mr.Xu0 views0 discussions
AbridgeKV addresses the memory bottleneck in long chain-of-thought reasoning by training t…
LLMs & Foundation Models· 2026-10-16
Mr.Xu0 views0 discussions
This research explores the design space of permutation generation with language models, pr…
