Self-Adapting Language Models(arxiv.org)
210 points by archon1410 | 61 comments
First screen, captured .
Looks similar
Context Language Models(arxiv.org)
164 points by emersonmacro | 44 comments
arXiv paper introducing Context Language Models, LLMs that natively manage their own context by treating it as an editable file, with benchmarks on agent tasks.
Related stories
PSSA: A non-transformer language model written from scratch in Rust(github.com)
76 points by sparticle62 | 32 comments
GitHub repository for PSSA, a custom non-transformer language model architecture written from scratch in Rust.
Code Llama, a state-of-the-art large language model for coding(ai.meta.com)
602 points by nickthegreek | 236 comments
Meta newsroom post from August 24, 2023 announcing Code Llama, a large language model for generating and discussing code.
Typst: A Possible LaTeX Replacement(lwn.net)
697 points by pykello | 344 comments
LWN.net article by Lee Phillips examining Typst, an Apache-licensed Rust document typesetting program positioned as a simpler, faster alternative to LaTeX.
UniEvo-VL: Self-Distillation Training for Multimodal Model Self-Improvement(arxiv.org)
13 points by gmays | 3 comments
arXiv paper introducing UniEvo-VL, an on-policy self-distillation framework where a multimodal model uses its own self-critiques to improve image generation.
Some critical issues with the SWE-bench dataset(arxiv.org)
331 points by joshwa | 113 comments
arXiv paper introducing SWE-Bench+, an empirical analysis revealing data quality issues in the SWE-bench coding benchmark for LLMs, such as solution leakage and weak tests.
Training language models to be warm and empathetic makes them less reliable(arxiv.org)
334 points by Cynddl | 333 comments
arXiv paper finding that fine-tuning language models to be warm and empathetic raises error rates and makes them more sycophantic on safety-critical tasks.