Highlights from UTC Meeting #184
blog.unicode.org·13h
🌏Character Sets
Accelerating Offensive R&D with Large Language Models
outflank.nl·4h
🔒WASM Capabilities
Show HN: Empromptu.ai – No code, AI app builder with RAG, model, evals etc.
news.ycombinator.com·1d·
Discuss: Hacker News
🌀Brotli Internals
Categorising My Daily Todo List with Deepseek-R1
bentasker.co.uk·6h·
Discuss: Hacker News
🌳Incremental Parsing
Parallel PLL on DAGs
arxiv.org·1d
🔗Topological Sorting
The Mystery of the Claude 3 Tokenizer (2024)
tokencontributions.substack.com·2d·
Discuss: Substack
🧪Binary Fuzzing
Survey of NLU Benchmarks Diagnosing Linguistic Phenomena: Why not Standardize Diagnostics Benchmarks?
arxiv.org·2d
⚙️Compression Benchmarking
VizGenie: Toward Self-Refining, Domain-Aware Workflows for Next-Generation Scientific Visualization
arxiv.org·1d
🏠Homelab Archaeology
Distribution-Based Masked Medical Vision-Language Model Using Structured Reports
arxiv.org·1d
📄OCR
Beyond Neural Networks: Symbolic Reasoning over Wavelet Logic Graph Signals
arxiv.org·1d
🌳Context free grammars
How I Fine-Tuned Granite-Vision 2B to Beat a 90B Model — Insights and Lessons Learned
towardsdatascience.com·5d
🌀Brotli Internals
GitHub Copilot: Reusable Prompts, Part 2 — Walkthrough & Real Examples
dev.to·1d·
Discuss: DEV
⚔️Lean Tactics
Enhancing Spatial Reasoning through Visual and Textual Thinking
arxiv.org·2d
🤖Advanced OCR
SeeDiff: Off-the-Shelf Seeded Mask Generation from Diffusion Models
arxiv.org·2d
🕵️Vector Smuggling
From Sufficiency to Reflection: Reinforcement-Guided Thinking Quality in Retrieval-Augmented Reasoning for LLMs
arxiv.org·11h
🔍Information Retrieval
React Libraries for Building AI-Powered Apps
dev.to·3h·
Discuss: DEV
🎙️Whisper
Should You Validate Again? A Smarter Way to Handle Data Trust
hackernoon.com·1d
✅Format Verification
RoD-TAL: A Benchmark for Answering Questions in Romanian Driving License Exams
arxiv.org·2d
🤖Automated Parsing
Where to show Demos in Your Prompt: A Positional Bias of In-Context Learning
arxiv.org·11h
💻Programming languages
Why not? Developing ABox Abduction beyond Repairs
arxiv.org·1d
🔲Proof Irrelevance