Boda Sadallah

We are releasing an expanded version of Global PIQA! It now covers 141 language varieties and includes parallel and non-parallel splits. We are also releasing an updated preprint.

@AndrewYNg @jyx_su Wanna evaluate the utilization of peer review comments written by human or AI reviewers? Use our system published at the last EMNLP: aclanthology.org/2025.emnlp-mai…

Releasing a new "Agentic Reviewer" for research papers. I started coding this as a weekend project, and @jyx_su made it much better. I was inspired by a student who had a paper rejected 6 times over 3 years. Their feedback loop -- waiting ~6 months for feedback each time -- was painfully slow. We wanted to see if an agentic workflow can help researchers iterate faster. When we trained the system on ICLR 2025 reviews and measured Spearman correlation (higher is better) on the test set: - Correlation between two human reviewers: 0.41 - Correlation between AI and a human reviewer: 0.42 This suggests agentic reviewing is approaching human-level performance. The agent grounds its feedback by searching arXiv, so it works best in fields like AI where research is freely published there. It’s an experimental tool, but I hope it helps you with your research. Check it out here: paperreview.ai

RevUtil includes: • 1.4k human-labeled + 10k synthetic review comments • Models (incl. 8B open) matching large closed LLMs • New benchmark for evaluating review utility Paper: aclanthology.org/2025.emnlp-mai… Code: github.com/bodasadallah/R… #NLProc (2/2)

We just presented RevUtil at EMNLP — the first dataset + system for automatically judging how useful peer-review feedback is. Why it matters: submissions are skyrocketing, but good, actionable reviews aren’t. (1/2)

Introducing Global PIQA, a new multilingual benchmark for 100+ languages. This benchmark is the outcome of this year’s MRL shared task, in collaboration with 300+ researchers from 65 countries. This dataset evaluates physical commonsense reasoning in culturally relevant contexts.

🔴 إيجاز هام | حول حالة السفن بشكل خاص • تم تأكيد اعتراض وقرصنة 22 سفينة، مع تقدير أن 19 سفينة أخرى قد تم اعتراضها دون توثيق. • من أصل 4 سفن متبقية: • سفينتان ابتعدتا لكونهما سفن دعم. • سفينة “مارينيت” ما زالت تبحر باتجاه غزة لكنها بعيدة بعد التحاق متأخر بسبب أعطال فنية. • السفينة الأخيرة “ميكينو” ظهرت على مواقع التتبع داخل المياه الإقليمية الفلسطينية على بعد 8 أميال من غزة، في حدث غير مسبوق منذ 20 عامًا من الحصار. • لم يتأكد بعد إن كان وصولها فعليًا وتوقفها هناك، أو نتيجة خلل تقني، أو اقتيادها بعد اعتراضها. • في حال تأكد وصولها، فسيكون حدثًا استثنائيًا يضاف لإنجازات الأسطول، الذي تمكنت سفنه من الوصول لأماكن لم يسبق لسفينة كسر حصار وصولها فيما مضى. #GlobalSumudFlotilla⁩ #أسطول_الصمود_العالمي

IMAGINE SENDING 20 WARSHIPS TO INTERCEPT BOATS CARRYING MILK POWDER AND BABY FORMULA FOR STARVING CHILDREN. THIS IS THE GREATEST EVIL EXISTING IN THE WORLD RIGHT NOW!!!!!

سألنى صديق لى ما الفائده من الدوران حول الكعبه فوجدت الاجابه فى كتاب العلامه الدكتور مصطفى محمود حوار مع صديقى الملحد وهو يوضح كيف أن مناسك الحج ليست خرافة ولا وثنية، بل هي انعكاس لقوانين الكون والطبيعة، وجزء من النظام الذي صنعه "الله عز وجل "...

الجيش رن على الشاب محمد الطباسي طلب منه يقف خلف تلة رمل بمنطقة سوافي خانيونس خلف مخيم الرحمة للنازحين حتي يقصفوه لوحده وإذا رفض سيتم استهداف عائلات المخيم و فعلا ودع اهله و قلهم ( خلوا الناس تدعيلي وتسامحني ) و صار يجري وتوجه خلف تلة رمل وكبرً عشان يصلي وقصفًوه واستشهد

what software engineers used to say: "it depends" what software engineers say now: "still not working pls fix"

RT @Villgecrazylady: They’re blowing up children so this demonic woman and her band of merry psychopaths can continue pursuing their Greate…

Some people are ready to lose their job/career for justice! Thanks to Ibtihal Aboussad who interrupted @Microsoft AI CEO @mustafasuleyman to protest MS support to the criminal occupiers of Israel in their genocide against the Palestinians! #FreePalestine

I have just had my first reviewing experience, and I thought I would share some insights about my experience. I have created a short blog post about it! Post link: bodasadallah.github.io/blog/first_rev… #NLP #Reserach #peer_review

Fixed it.

This is Hossam’s team, and we are sharing his final message : “If you’re reading this, it means I have been killed—most likely targeted—by the Israeli occupation forces. When this all began, I was only 21 years old—a college student with dreams like anyone else. For past 18 months, I have dedicated every moment of my life to my people. I documented the horrors in northern Gaza minute by minute, determined to show the world the truth they tried to bury. I slept on pavements, in schools, in tents—anywhere I could. Each day was a battle for survival. I endured hunger for months, yet I never left my people’s side. By God, I fulfilled my duty as a journalist. I risked everything to report the truth, and now, I am finally at rest—something I haven’t known in the past 18 months . I did all this because I believe in the Palestinian cause. I believe this land is ours, and it has been the highest honor of my life to die defending it and serving its people. I ask you now: do not stop speaking about Gaza. Do not let the world look away. Keep fighting, keep telling our stories—until Palestine is free.” — For the last time, Hossam Shabat, from northern Gaza.

Update: We have submitted another 23GB of raw war crimes footage (around 10,000 videos & images), to the @IntlCrimCourt South African, @HindRFoundation, and @ICJPalestine legal teams for ongoing cases. Thanks to our research group for tirelessly documenting these war crimes for the past 520 days.

🚨BREAKING: Five medical staff, nurses, and wounded patients have been killed, and several others injured in an Israeli airstrike targeting the emergency building at Nasser Medical Complex in Khan Yunis, southern Gaza. The death toll is preliminary and expected to rise.

My position paper “LLM Alignment for the Arabs: A Homogenous Culture or Diverse Ones?” got accepted to the @c3_nlp workshop co-located with @naaclmeeting. I share concerns about the missed opportunities with the rise of Arabic-specific LLMs. 📜arxiv.org/abs/2503.15003 (1/4)