Expert attention map

The Who's Who of AI

What credible people across AI noticed, why it matters, and where the field is converging or disagreeing.

2,397 searchable experts 4,280 tracked across all sources
Build my stream Create my edition
Filter the conversation Who is saying what?

Combine a professional role with a reaction lens. Both must match the same attributed contribution.

Clear all
Active evidence filter

Showing developments with attributable Research & technical analysis reactions.

New Network reaction maps

See how experts are reacting—not just what they shared.

Posts are grouped by conversation and tone. Select a lens to filter the stream; these are never permanent labels on people.

Research & technical analysis
2 live reaction maps in this view Jump to the evidence ↓
Shared emphasis Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach Research & technical analysis · 2 attributable reactions Reaction converging Muse Wonder Research & technical analysis · 3 attributable reactions
Showing signals surfaced by Chris Paxton ×
Most important

The developments commanding sustained expert attention

One card per development. Sources are clustered; reaction bundles describe these posts, never the people behind them.

Established Compute & infrastructure Research 18d ago
⚡ 6 h early
latent reasoning recurrent depth test-time compute paper

Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach

3 experts across 3 network communities independently surfaced this.

Why this matches Research & technical analysis reaction 2 attributable expert contributions · Andreas Kirsch, Sung Kim
“This work builds on LCKV (final-layer caches + KV matching), CLA (cross-layer reuse), and Huginn (recurrent depth), but we test cache-history mismatch and repair in a tied core using LSTM cells aclanthology.org/2024.acl-lo... arxiv.org/abs/2405.12981 arxiv.…” evidence ↗
3 experts 3 communities 1 sources clustered
Andreas Kirsch Chris Paxton Sung Kim
How the network is reacting 2 experts are emphasizing research & technical analysis.
Shared emphasis

Research & technical analysis

2 experts

Evidence, methods and technical implications.

“OpenAI’s Astra may be using Recurrent Depth as outlined in this paper: Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach (arxiv.org/abs/2502.05171)”

Primary source arxiv.org ↗
Established AI field signal Development 11d ago
⚡ 2 h early
Meta Muse Wonder model launch

Muse Wonder

4 experts across 3 network communities independently surfaced this.

Why this matches Research & technical analysis reaction 3 attributable expert contributions · Nathan Lambert, Tim Kellogg, Sung Kim
“Looks like Meta is back at the frontier with their latest "Muse Wonder" models. Will be interesting to see independent evaluations / user testing. research.meta.ai/blog/muse-wo...” evidence ↗
4 experts 3 communities 1 sources clustered
Nathan Lambert Tim Kellogg Chris Paxton Sung Kim
How the network is reacting 3 experts are independently emphasizing research & technical analysis.
Reaction converging

Research & technical analysis

3 experts

Evidence, methods and technical implications.

“Looks like Meta is back at the frontier with their latest "Muse Wonder" models. Will be interesting to see independent evaluations / user testing. research.meta.ai/blog/muse-wo...”

2 experts discussed this · 5 posts
Sung Kim: Meta's Muse Wonder Muse Wonder = Muse Wonder Flash Muse Wonder Full = Muse Wonder so confusing... research.meta.ai/blog/muse-wo...
Nafnlaus 🇮🇸 🇺🇦: All for the pun...
Nafnlaus 🇮🇸 🇺🇦: Not sure how much I'd trust that.. my experience with Spark 1.3 and Spark 1.3 contributor has been... subpar vs. what the benchmarks suggest
Open the full discussion →
Primary source research.meta.ai ↗
Established Agents & robotics Analysis 11d ago
IROS 2026 robotics dexterity essay

Dexterity is the Frontier: Thoughts on IROS 2026

1 directory member surfaced this signal.

Why this matches Research & technical analysis reaction 1 attributable expert contribution · Chris Paxton
“Some thoughts on a robotics conference: the research community increasingly is looking at enabling contact-rich manipulation with dexterous hands, and has the tools to do so itcanthink.substack.com/p/dexterity-...” evidence ↗
1 expert 1 community 1 sources clustered
Chris Paxton

“Some thoughts on a robotics conference: the research community increasingly is looking at enabling contact-rich manipulation with dexterous hands, and has the tools to do so itcanthink.substack.com/p/dexterity-...”

Primary source itcanthink.substack.com ↗
Debates gaining attention

What experts are discussing without an anchoring article

3 experts · 3 posts · 6d ago
Eugene Vinitsky Chris Paxton jake
Matches Research & technical analysis
“The solution to interpretability is clearly the same as the solution to ai coding: just ask deepseek if claude did what it said it did, they can't BOTH be lying to you” evidence ↗
Multiple readings Research & technical analysis · 1 Questions & unknowns · 1
Chris Paxton: The solution to interpretability is clearly the same as the solution to ai coding: just ask deepseek if claude did what it said it did, they can't BOTH be lying to you
Eugene Vinitsky: Okay fine they both lied but if I ask 3 models it’s all good then we can do majority vote
Open the thread →