Clem Bonnet (@ClementBonnet16)

Pinned

C

Clem Bonnet@ClementBonnet16 · Feb 19

I had a fantastic time discussing latent search with @ecsquendor on MLST! 🙏 We review our work on latent program networks co-authored with @MattVMacfarlane and debate composition, AI creativity, and program synthesis. Full video: youtu.be/PHBItVuudbU

MMachine Learning Street Talk@MLStreetTalk · Feb 18

We spoke with @ClementBonnet16 at NeurIPS about his extremely innovative approach to the @arcprize using a form of test time inference where you search a latent space of a VAE before making an optimal prediction. @fchollet was so impressed, he hired Clem shortly after! 😃 -…

3

4

52

12

6.0K

Clem Bonnet Retweeted

A

ARC Prize@arcprize · Jul 18

Today, we're announcing a preview of ARC-AGI-3, the Interactive Reasoning Benchmark with the widest gap between easy for humans and hard for AI We’re releasing: * 3 games (environments) * $10K agent contest * AI agents API Starting scores - Frontier AI: 0%, Humans: 100%

65

225

2.0K

544

335.0K

Clem Bonnet Retweeted

S

Shao-Hua Sun@shaohua0116 · Jul 17

Our #ICML2025 Programmatic Representations for Agent Learning workshop will take place tomorrow, July 18th, at the West Meeting Room 301-305, exploring how programmatic representations can make agent learning more interpretable, generalizable, efficient, and safe! Come join us!

1

15

63

25

33.0K

Clem Bonnet Retweeted

A

Axel Darmouni@ADarmouni · Jul 15

Encoder-Decoder for Latent Program Search : how there was (and still is) a possibility to have specialized approaches for Arc problem solving 📖 Read 203: « Searching Latent Program Spaces », by @ClementBonnet16 and @MattVMacfarlane arxiv.org/pdf/2411.08706 A very original…

2

1

13

3

538

Clem Bonnet Retweeted

M

Matthew Macfarlane@MattVMacfarlane · Jul 10

I’ll be presenting two workshop papers: “Searching Latent Program Spaces” (oral at Programmatic Representations for Agent Learning) & “Instilling Parallel Reasoning into Language Models” (AI4Math).

1

9

2

364

Clem Bonnet Retweeted

L

Levi Lelis@levilelis · Jul 2

Previous work has shown that programmatic policies—computer programs written in a domain-specific language—generalize to out-of-distribution problems more easily than neural policies. Is this really the case? 🧵

1

6

30

4

2.0K

Clem Bonnet Retweeted

N

Ndea@ndea · May 9

Ndea is sponsoring SYNT 2025 - a workshop on synthesis of computing systems - July 22 in Zagreb, Croatia. Part of CAV 2025 (Conference on Computer Aided Verification). Have a synthesis-related paper abstract? Submissions due by May 18. synt2025.github.io

7

10

43

21

38.0K

C

Clem Bonnet@ClementBonnet16 · Apr 17

I will be at ICLR next week. Always up for chatting about ARC-AGI, program synthesis, RL, open-endedness, or related rabbit holes. DMs open! #ICLR2025

1

25

3

1.0K

Clem Bonnet Retweeted

F

François Chollet@fchollet · Apr 17

Compressing the timeline to get to AGI also means compressing the timeline of every single scientific breakthrough that is downstream of AGI. There is no greater leverage

47

57

712

95

61.0K

Clem Bonnet Retweeted

L

Lewis Hemens@lewishemens · Apr 7

I've just finished drafting a fairly thorough review of a lot of the research that went into @arcprize 2024, all the paper winners, top scorers, and a few hot takes for ARC-AGI-2 in 2025. Here's a short 🧵 with some highlights!

6

14

88

68

25.0K

Clem Bonnet Retweeted

N

Ndea@ndea · Mar 27

Quick hiring update: we've assembled an incredible founding research team. We have no open positions, for now. However — we will create roles for exceptional program synthesis researchers. If that's you: ndea.com/join Onward.

1

2

26

6

5.0K

Clem Bonnet Retweeted

A

ARC Prize@arcprize · Mar 26

ARC Prize 2025 is Live $1M competition to open source a solution to ARC-AGI Your objective: Reach 85% on the private evaluation dataset Progress needs new ideas, not just scale

7

33

163

22

38.0K

Clem Bonnet Retweeted

A

ARC Prize@arcprize · Mar 24

Today we are announcing ARC-AGI-2, an unsaturated frontier AGI benchmark that challenges AI reasoning systems (same relative ease for humans). Grand Prize: 85%, ~$0.42/task efficiency Current Performance: * Base LLMs: 0% * Reasoning Systems: <4%

69

337

2.0K

641

453.0K

Clem Bonnet Retweeted

M

Mohamed Osman@MohamedOsmanML · Mar 23

Honored to be a guest on the infamous MLST podcast again! We discuss our test-time methods, compositionality in LLMs, limitations of VLMs, logic vs perception, efficient adaptation, and more. @MLStreetTalk youtu.be/3p0O28W1ZHg

6

9

21

8

5.0K

Clem Bonnet Retweeted

A

ARC Prize@arcprize · Mar 14

3/24/2025

12

22

313

34

75.0K

Clem Bonnet Retweeted

N

Nathan Grinsztajn@NGrinsztajn · Mar 13

So proud of this one: today we're releasing Command A, our new 111B flagship model tailored for business use cases. Gpt4o/Deepseek-level in evals, better than Sonnet on LMsys, has 256k context length. Also, weights are available now on hf! huggingface.co/CohereForAI/c4…

1

7

50

3

2.0K

Clem Bonnet Retweeted

N

Ndea@ndea · Feb 27

Deep Learning architectures usually aren't trained to perform search at test time, leading to sample inefficiency + poor generalization. Latent Program Network (LPN) builds in test-time adaption by learning a latent space that can be searched. @ClementBonnet16 @MattVMacfarlane

5

20

205

144

30.0K

Clem Bonnet Retweeted

A

ARC Prize@arcprize · Feb 25

AGI is reached when the capability gap between humans and computers is zero ARC Prize Foundation measures this to inspire progress Today we preview the unbeaten ARC-AGI-2 + open public donations to fund ARC-AGI-3 TY Schmidt Sciences (@ericschmidt) for $50k to kick us off!

24

67

679

133

222.0K

C

Clem Bonnet@ClementBonnet16 · Feb 18

Checkout @ClementBonnet16 discussing searching latent program spaces tommorow on @MLStreetTalk !

MMachine Learning Street Talk@MLStreetTalk · Feb 18

We spoke with @ClementBonnet16 at NeurIPS about his extremely innovative approach to the @arcprize using a form of test time inference where you search a latent space of a VAE before making an optimal prediction. @fchollet was so impressed, he hired Clem shortly after! 😃 -…

0

2

25

1

2.0K

Clem Bonnet Retweeted

M

Machine Learning Street Talk@MLStreetTalk · Feb 18

We spoke with @ClementBonnet16 at NeurIPS about his extremely innovative approach to the @arcprize using a form of test time inference where you search a latent space of a VAE before making an optimal prediction. @fchollet was so impressed, he hired Clem shortly after! 😃 -…

8

35

281

160

37.0K