SO-Bench: A Structural Output Evaluation of Multimodal LLMs
Multimodal large language models (MLLMs) are increasingly deployed in real-world, agentic settings where outputs must…
Recent Articles
View All
On the Aesthetic of Wizard Power
Published on December 4, 2025 11:18 PM GMTEpistemic status: A response to @johnswentworth's "Orienting Towards…
Micro1, a Scale AI competitor, touts crossing $100M ARR
Micro1 started the year with roughly $7 million ARR. Now, it claims to have surpassed…
Will misaligned AIs know that they’re misaligned?
Published on December 4, 2025 9:58 PM GMTEpistemic status: exploratory, speculative.Let’s say AIs are “misaligned”…
Anthropic CEO weighs in on AI bubble talk and risk-taking among competitors
Anthropic's CEO shared his thoughts on the economics of AI and the risk-taking of competitors,…
Thresholding
Published on December 4, 2025 7:53 PM GMT(This is a linkpost for Duncan Sabien's article…
An Abstract Arsenal: Future Tokens in Claude Skills
Published on December 4, 2025 8:01 PM GMTtl;drDimensionalize. Antithesize. Metaphorize. These are cognitive tools in…
AI chatbots can sway voters better than political advertisements
In 2024, a Democratic congressional candidate in Pennsylvania, Shamaine Daniels, used an AI chatbot named…
Cross Layer Transcoders for the Qwen3 LLM Family
Published on December 4, 2025 7:11 PM GMTDigging Into Interpretable FeaturesSparse autoencoders SAEs and cross…