I trained a small transformer in 1.5hrs and it beats many LLMs
(No description available)
→ View original source
DramaChain Bench: An End-to-End Benchmark for Short-Drama Generation
DramaChain Bench introduces an end‑to‑end benchmark for short‑drama generation that spans the full production pipeline—from script and storyboard to keyframe imagery, shot‑level video, and final drama—rather than focusin…
→ View original source
From Production Traffic to Post-Training: Building a Self-Hosted LLM That Covers the Corporate Request Mix
Enterprises need to self‑host large language models to meet data‑residency requirements, yet supporting diverse internal applications fragments a limited GPU pool. The authors consolidate traffic from over 200 internal s…
→ View original source
Control-Data Flow Separation: Stable Prompt Optimization in Multi-Agent LLMs
Prompt optimization can improve multi-agent LLM systems, but the prompts being optimized often serve two entangled roles: generating task-relevant content and specifying execution-critical protocols,
→ View original source
Qwen-Drive-1.0: An Initial Step towards a Vision-Language Foundation Model for Autonomous Driving
We present Qwen-Drive-1.0, an initial step towards a vision-language foundation model for autonomous driving. Qwen-Drive-1.0 retains the architecture of the pretrained vision-language model (VLM) and
→ View original source
New Model: Spark-X2.5-4B, Spark-X2.5-1.7B
XHToken has released two new small language models, Spark-X2.5-4B and Spark-X2.5-1.7B, featuring a custom architecture rather than being fine-tunes of existing models. Both versions claim native 1M context window support…
→ View original source
The craziest thing about Fable 5.1 for me personally
A Reddit user analyzed two months of usage data for Fable 5.1 via ccusage, finding that cache read costs account for 78% of total Claude usage. If cache read pricing drops by 75% as expected, overall costs would decrease…
→ View original sourceThe ChatGPT/Codex app bundles a full copy of LibreOffice
(No description available)
→ View original source
Claude AI Prompt Engineering System (2026) — Write Better Prompts, Get Better Results
Claude AI's 2026 Prompt Engineering System
→ View original sourceImbad0202 /academic-research-skills
This repository offers a structured workflow for academic research using Claude Code, covering research, writing, review, revision, and finalization phases. It aims to streamline scholarly work through a systematic, iter…
→ View original source
Which LLMs will run on the Mac Mini and Studio
A mildly interesting video about which models can run on the Mac Mini and Studio. As expected current foundational models such as Kimi K3 1.4 TB, if they were even available to run lo
→ View original source
Researchers from Princeton, Ant Group and Stanford Introduce AQuA: A Two-Part Agentic Framework for Autonomous Factor Discovery and Model Development in Quantitative Finance
Researchers from Princeton, Ant Group and Stanford introduced AQuA, a two‑part agentic framework for autonomous factor discovery and model development in quantitative finance. The framework separates symbolic factor disc…
→ View original source