YY Ahn

As people have predicted, there are quite a few open-source alternatives to Jev, including: - Laya, this was the original according to him: github.com/NandhaKishor... - on Apple Core ML / neural engine: github.com/mizorewww/la... - with Qwen: github.com/jaredpalmer/...

GitHub - NandhaKishorM/laya github.com

As people have predicted, there are quite a few open-source alternatives to Jev, including: - Laya, this was the original according to him: github.com/NandhaKishor... - on Apple Core ML / neural engine: github.com/mizorewww/la... - with Qwen: github.com/jaredpalmer/...

GitHub - mizorewww/laya-coreml: Local Laya typed decisions on Apple Core ML and Neural Engine. Validated ports, ~5 ms short decisions on M3 Max, reproducible speed and energy benchmarks. github.com

As people have predicted, there are quite a few open-source alternatives to Jev, including: - Laya, this was the original according to him: github.com/NandhaKishor... - on Apple Core ML / neural engine: github.com/mizorewww/la... - with Qwen: github.com/jaredpalmer/...

github.com

- nanojev: github.com/TianyuCoding... - awesome-jev: github.com/yibie/awesom...

GitHub - TianyuCodings/NanoJev: A nano replica of Jev: parallel decisions, dynamic candidates, and an end-to-end training pipeline. github.com

- nanojev: github.com/TianyuCoding... - awesome-jev: github.com/yibie/awesom...

GitHub - yibie/awesome-jev: A curated list of public projects, integrations, and discussions built on Jev — TypeSafe AI's System One model for typed decisions. github.com

And Zhou et al. find that returns can eventually turn negative: at high budgets, models sometimes abandon an earlier correct answer. Easy problems peak much sooner than hard ones. arxiv.org/abs/2604.10739

When More Thinking Hurts: Overthinking in LLM Test-Time Compute Scaling arxiv.org

But Snell et al. found that the best inference strategy depends on how hard the problem is; a fixed budget is wasteful because easy and hard prompts benefit from different amounts of computation. arxiv.org/abs/2408.03314

Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters arxiv.org

Many test-time-compute papers show why this is hard. In an extreme version, just appending "Wait" whenever the model tried to stop raised the benchmark. Often, just forcing more reasoning does work: arxiv.org/abs/2501.19393

s1: Simple test-time scaling arxiv.org

I think that's exactly what many skills are about. Many of github.com/yy/claude-sc... are basically such checklists to go through.

GitHub - yy/claude-scholar: Academic research tools for Claude Code — literature search, citation management, LaTeX checks, math verification, and submission preparation github.com

Neural Scaling Universality: If Exponents Are Fixed, Time to Understand Coefficients arxiv.org/abs/2606.25008

Neural Scaling Universality: If Exponents Are Fixed, Time to Understand Coefficients arxiv.org

"Leak, Cheat, Repeat: Data Contamination and Evaluation Malpractices in Closed-Source LLMs" arxiv.org/abs/2402.03927

Leak, Cheat, Repeat: Data Contamination and Evaluation Malpractices in Closed-Source LLMs arxiv.org

This paper looks cool: arxiv.org/abs/2605.23901 "We propose the Shannon Scaling Law, a unified theoretical framework that models LLM training as information transmission over a noisy channel ... The Shannon Scaling Law consistently outperforms classical scaling laws ..."

LLMs as Noisy Channels: A Shannon Perspective on Model Capacity and Scaling Laws arxiv.org

Meta Muse, which is pretty well made, may be the beginning of the next-gen AI agent for the masses, but of course with potential privacy/surveillance nightmares.

"A lecture is the process whereby the notes of the professor become the notes of the student without passing through the mind of either." I think this may apply to lots of "LLM wiki" usage.

I'm suspecting a major fault line in AI discourse may hinge upon where one stands on Searle’s Chinese Room argument.