/
Navigation
C
Chronicles
Browse all articles
C
E
Explore
Semantic exploration
E
R
Research
Entity momentum
R
N
Nexus
Correlations & relationships
N
~
Story Arc
Topic evolution
S
Drift Map
Semantic trajectory animation
D
P
Posts
Analysis & commentary
P
Browse
@
Entities
Companies, people, products, technologies
Domains
Browse by publication source
Handles
Browse by social media handle
Detection
?
Concept Search
Semantic similarity search
!
High Impact Stories
Top coverage by position
+
Sentiment Analysis
Positive/negative coverage
*
Anomaly Detection
Unusual coverage patterns
Analysis
vs
Rivalry Report
Compare two entities head-to-head
/\
Semantic Pivots
Narrative discontinuities
!!
Crisis Response
Event recovery patterns
Connected
Nav: C E R N
Search: /
Command: ⌘K
Embeddings: large
VOICE ARCHIVE

Luca Soldaini

@soldni
4 posts
2026-01-28
scaling laws for agent specialization are so cool-unsurprisingly, you can get much better multipliers if you do repo specific trajectory generation great work @Tim_Dettmers @ethnlshn! [image]
2026-01-28 View on X
SiliconANGLE

Ai2 launches Open Coding Agents, starting with SERA, an open-source family that includes 32B and 8B parameter models designed to adapt to private codebases

2026-01-27
scaling laws for agent specialization are so cool-unsurprisingly, you can get much better multipliers if you do repo specific trajectory generation great work @Tim_Dettmers @ethnlshn! [image]
2026-01-27 View on X
SiliconANGLE

Ai2 launches Open Coding Agents, starting with SERA, an open-source family that includes 32B and 8B parameter models designed to adapt to private codebases

Artificial intelligence is moving swiftly, changing how developers craft, as code flows ever faster into repositories such as GitHub …

2025-12-16
I've been grumbling about tokenizers but it took a @bminixhofer to do something about it! Really neat approach: rather than committing to tokenizer-free methods from the get-go, we show how to switch from BPE at any point during training run 👾
2025-12-16 View on X
VentureBeat

Allen Institute for AI launches Bolmo 7B and Bolmo 1B, claiming they are “the first fully open byte-level language models”, built on its Olmo 3 models

and every token gets the same compute, regardless of complexity. Benjamin Minixhofer / @bminixhofer : There are also some things Bolmo lets us do which we just can't do using subwo...

2023-06-27
very good & timely article by @jjvincent beside harms to minortitized populations, bland AI-generated content polluting the internet is one of the saddest effects of current wave of LLMs / image generation models https://www.theverge.com/...
2023-06-27 View on X
The Verge

AI is killing the web with spam sites, mass article producing “AI editors”, AI-generated junk on Etsy, Reddit, and Wikipedia that exhausts moderators, and more

In recent months, the signs and portents have been accumulating with increasing speed.  Google is trying to kill the 10 blue links.