Nir Mazor

[Nir Mazor @NirMMazor]

New preprint 💥

Can a general-purpose model achieve results comparable to medically pre-trained models? 🤔

We show that lightweight fine-tuning of a general-purpose LVLM and an LVLM-aware retriever can. 🚀

🔗 GitHub: https://t.co/vbHBIyoNgj

📄 Paper: https://t.co/Q8tKLk8J1P

[Noy Sternlicht @NoySternlicht]

🎉 Happy to share that CHIMERA has been accepted to #ACL2026NLP (main conference)!

📄 Paper - https://t.co/11tMhA07P2
🤗 Data - https://t.co/g8GgYsfj9E
🌐 Project Page - https://t.co/c4j8gEm6V5
💻 Git - https://t.co/lVC8UAzOti

Joint work with @Hoper_Tom

[Eliya Habba @EliyaHabba]

New datasets keep coming,
New models keep coming.

Frustrating!
How can we evaluate everything on everything?
How do we keep scores comparable over time?

We propose a way to grow benchmark suites without losing comparability.

Details:👇🧵

[Michael Hassid @MichaelHassid]

Have LLMs become supervised learners (once again)?!

In our new paper, we argue that current LLMs’ post-training methods have effectively reverted to the "pre-train then fine-tune" era, explicitly tailoring models to desired behaviors.

1/n

[Guy Kaplan @GKaplan38844]

Fine-Tuning LLMs on New Knowledge Encourages Hallucinations. (@zorikgekhman)
But why? We found something unexpected:
1M facts about city-like names →hallucinations explode.
1M facts about random identifiers →near zero!
Same model. Same number of facts. Only the names change.🧵

[Itay Itzhak @Itay_itzhak_]

Ever used a top-ranked LLM that just... felt wrong for you?

You’re not alone. Instead of leaderboards, many of us turn to "vibe-testing" - manually comparing models to our own needs. But can we turn these feelings into a structured evaluation?

New paper: "From Feelings to Metrics" 🧵

[Roni Itkin @ItkRoni]

We introduce 🌍GlobalSplat: Efficient Feed-Forward 3D Gaussian Splatting via Global Scene Tokens.🌍

Most feed-forward 3DGS methods still start from pixel, voxel, or dense view-aligned primitives.
We take a different route: align first, decode later. 🧵👇

[Ariel Goldstein @GoldsteinYAriel]

We are excited to share our new preprint! 🚨
https://t.co/mRKThbiSbH
In this work, we investigate the neural mechanisms by which speakers generate information-rich linguistic content during spontaneous natural conversation.

[Eliya Habba @EliyaHabba]

What if you could automatically turn a large collection of documents into structured databases, tailored for your own research needs?📄

We introduce ScheMatiQ!
From question ➝ schema ➝ structured data 🔍

@ShaharLevy19 @MintzReshef @RKeydar @BarakRaveh @GabiStanovsky
🧵
👇1/5

[Tom Hope @Hoper_Tom]

Happy to share three papers accepted to the #ACL2026 @aclmeeting conference: on AI ideation, multimodal medical AI, and literature understanding.

If you want to read about them, check out the following posts from @NoySternlicht , @NirMMazor and @UKPLab . More updates on each to come separately.

In the hypothesis generation / ideation space : cross-domain inspirations. Led by @NoySternlicht .
https://t.co/d91iDkZ4B2

In the multimodal medical AI domain: medical image diagnosis with literature retrieval. Led by @NirMMazor .
https://t.co/hVRGITKfwM

In literature understanding space: summarizing thousands of citations to understand a paper's impact. Led by Hiba Arnaout from @IGurevych's lab, with major contributions from @NoySternlicht .
https://t.co/ZTWZhbEWZX

[Noam Dahan @Dahan_Noam]

Today at #EACL2026 we present our work exploring PRINTED newspapers as a data source for summarization in low-resource languages.
Join the virtual session at 18:00!
🧵👇

[Shachar Don-Yehiya @Shachar_Don]

Do you run pairwise evaluation?
Do you test your models on the Arena-Hard and AlpacaEval benchmarks?

You probably want to read this 🧵👇

https://t.co/z5gxRdXHHb
With @LChoshen @AbendOmri

[Asaf Yehudai @AsafYehudai]

New preprint, evaluation framework & leaderboard!🚨

General-purpose AI agents are everywhere. 🤖
From ReAct to @claudeai Code and @OpenAI SDK.

But how do we actually evaluate them — as general agents?

Currently, benchmarks are deeply tied to domain-specific setups, making it impossible to evaluate true cross-domain agents.

We’re changing that!

We’re introducing Exgentic and the Open General Agent Leaderboard. 🧵👇

[Oren Sultan @oren_sultan]

Can LLMs reliably predict program termination?

We evaluate frontier LLMs in the International Competition on Software Verification (SV-COMP) 2025, directly competing with state-of-the-art verification systems.

@AIatMeta @HebrewU @Bloomberg @imperialcollege @ucl

@jordiae @pascalkesseli @jvanegue @HyadataLab @adiyossLC @PeterOHearn12

Paper: https://t.co/5GP6q8E87v
Website: https://t.co/HQwhl2UVLQ

🧵👇
1/n

[Nir Mazor @NirMMazor]

Under the guidance of @Hoper_Tom — thank you! @nlphuji

[Nir Mazor @NirMMazor]

New preprint 💥

Can a general-purpose model achieve results comparable to medically pre-trained models? 🤔

We show that lightweight fine-tuning of a general-purpose LVLM and an LVLM-aware retriever can. 🚀

🔗 GitHub: https://t.co/vbHBIyoNgj

📄 Paper: https://t.co/Q8tKLk8J1P

[Nir Mazor @NirMMazor]

Our model also achieves superior results over general-purpose RAG baseline models 🚀📈

[Avishai Elmakies @AvishaiElm37946]

🚀 Excited to share that my paper from my internship at @IBMResearch has been accepted to #ICASSP2026!
We train Speech-Aware LLMs (SALLMs) with Group Relative Policy Optimization (GRPO) on open-ended tasks (Spoken QA & Speech Translation). We find that GRPO beats SFT!