@HamriMay
 Israel
Joined May 2023
  • Tweets12
  • Following29
  • Followers10
1Media
Accepted to #NeurIPS2026! 🎉
may hamri@HamriMayJun 12
🦓 New paper: ZEBRA - Your AI agent has $10 to spend on a coding task across planning, implementing, and refining. How should it split the budget? We turn budget allocation into a knapsack problem, solved zero-shot at inference time.With @InbalTalgam 👇 [arxiv.org/abs/2605.20485]

ZEBRA: Zero-shot Budgeted Resource Allocation for LLM Orchestration

As autonomous agents increasingly execute end-to-end tasks under fixed monetary budgets, the pressing open question shifts from whether the budget is respected, to how to spend it effectively....

arxiv.org
3
2
12
1,989
You ask for a talking parrot, but the model makes the pirate talk AND gives him a beak! 🦜😮‍💨 -> Introducing The Attention Triangle in Audio-Video Models: how semantics leak between text, audio, and video, and how we steer them back. 🧵 [1/5]
3
27
1
60
1,961
Robot policies can move but can't think. LLMs can think but can't move. So we connected them. Real robot: 16.7% → 97.3% Sim (LIBERO-PRO): 12.8% → 53.3%
23
54
9
305
96,263
Looking forward to presenting ZEBRA at the SCALE Workshop @icmlconf next month!🦓 If you’ll be at ICML, come say hi!
2
76
ZEBRA recover 94% of unconstrained quality at half the spend, beating direct LLM allocation methods on every metric.
1
3
122
🦓 New paper: ZEBRA - Your AI agent has $10 to spend on a coding task across planning, implementing, and refining. How should it split the budget? We turn budget allocation into a knapsack problem, solved zero-shot at inference time.With @InbalTalgam 👇 [arxiv.org/abs/2605.20485]

ZEBRA: Zero-shot Budgeted Resource Allocation for LLM Orchestration

As autonomous agents increasingly execute end-to-end tasks under fixed monetary budgets, the pressing open question shifts from whether the budget is respected, to how to spend it effectively....

arxiv.org
4
1
2
9
2,707
Gave a talk at @GoogleDeepMind on our joint work with @HazanPrinceton and @tri_dao on deployable RL policies for robotics. Great discussion with @JeffDean.
10
19
3
172
23,908
[1/4] Sync about it… 💭✨ Editing a portrait video yet keeping it fully synced with the original across the entire sequence. Read more about Sync-LoRA: sagipolaczek.github.io/Sync-… 🚀
2
20
3
46
6,828
Feels like a dream! I’ve recently started my Ph.D. in Computer Science @Princeton! Working on exciting research with Professors @HazanPrinceton and @tri_dao 🤩
107
90
15
2,234
175,840
[1/5] Rethinking SVGs, the implicit way — meet NeuralSVG! 🎨✨ An implicit neural representation for generating layered SVGs from text prompts. 💧 Powered by SDS and nested-dropout for ordered shapes 🖌️ Enables inference-time editing like color palette & aspect ratio Read more on sagipolaczek.github.io/Neura… 🚀
8
52
4
296
27,928
may hamri retweeted
Roi Cohen
@roicohen923 May 2023
LLMs make factual mistakes, which is bad. How can we find these? In a new paper, we introduce “LMvsLM”, and show that we can use one LM to detect the mistakes of another via "Cross-Examination". Link: arxiv.org/abs/2305.13281 Check it out!
4
18
2
78
11,183

No more items