Abhinav Java

Incoming PhD student @mldcmu | Prev - RF @MSFTResearch, MLE @Adobe | @dtu_delhi | Hala Madrid!

[Abhinav Java @abhinav_java]

Our #CVPR 2026 paper is selected for an oral presentation🌟

One of my favorite projects as a Research Fellow at MSR India w @SachdevaBhuvan, @karan_uppal3, and Vineeth NB!

@SachdevaBhuvan is extremely driven and talented. If you’ve got RE roles, reach out now!

[Bhuvan Sachdeva @SachdevaBhuvan]

Our paper has been accepted for Oral presentation at #CVPR 🎉🎉
Kudos to the team: @karan_uppal3, @abhinav_java
See you in Denver!

On a side note, I am finishing my fellowship by June and looking for full-time research roles. DMs open.

[Abhinav Java @abhinav_java]

@Ed_Acteson I swear I thought this is a sarcastic post, afraid not 😂

[Chinmay @ChinmayKak]

On a more personal note:
I am wrapping up my internship and looking for full-time roles in post-training and adjacent areas. At MSR I worked on agentic RL for general tool-calling on benchmarks like BFCL, Taubench, and MCPBench and on self-distillation and synthetic-environment data curation and training.
More of my work can be found on my github and website, linking below.
Please RT/QT for helping w the algo:)

[Raj Chohan @rajsinghchohan]

@RDrfx_RMCF Matty F is good defensively whilst providing passing range, Martinez is a muppet

[Abhinav Java @abhinav_java]

@swayaminsync You didn't feed him enough bro

[Abhinav Java @abhinav_java]

the black one completed its research fellowship at MSR India 🐾🐾

[Swayam Singh @swayaminsync]

The orange one did his first hunt today and man he got a pitch-perfect meow.
Proud!!

[Abhinav Java @abhinav_java]

Very insightful read.

Last year, we provided an alternative perspective - can we build model agnostic unlearnable datasets? (https://t.co/zvj0jhqHKo)

excited to explore how some of these ideas could help create private datasets!

[Gaurav Ghosal @gaurav_ghosal]

We are taking a big step towards scaling LLMs that can unlearn on demand.
Cleanly deleting data from LLMs has proven impossible: training entangles every source in shared weights.
NULLs (Natively Unlearnable LLMs) escapes this, keeping millions of sources individually deletable in a 1B-parameter model trained on web data. (1/8)

[Ashish Negi @7negiashish]

Schoolboy football from Khalid Jamil's India. No game awareness, no tactical structure & very little technical quality shown throughout the 90 minutes as India were thrashed by Tajikistan.

But don't worry, the preparation for the SAFF "World" Championship is apparently going very well. #IndianFootball

[Hanna Hajishirzi @HannaHajishirzi]

MAI-Thinking-1 is out!

Excited to share what we are building and how climbing from scratch (no distillation) actually works: simple recipes, rigorous science, self-distillation, patience, and great infra.

Check out our tech report has the full story of our RL climbs.
https://t.co/aLW40sWz4d

[Guowei Xu @Kevin_GuoweiXu]

🚀 How should LLMs sample on hard reasoning problems during post-training and inference where direct rollouts rarely produce a correct answer?

Best-of-N (e.g., GRPO) and tree search share two limitations:
🔻 Verification signals are sparse
🔻 Candidates stay within the model's own distribution

We introduce BES: Bidirectional Evolutionary Search — a search framework that couples forward candidate evolution with backward goal decomposition.
✅ Works for both post-training and inference.

[Deepti Ghadiyaram @deeptigp]

[1/4] The human eye doesn't process every single pixel of a video continuously—it focuses on what changes.

So why are our video AI models wasting compute on redundant frames?

Introducing Swift Sampling: a test-time technique inspired by the human visual system. 🧠👇

[Abhinav Java @abhinav_java]

Unexpected Claude Code hack: instead of letting it write the code, ask it to generate exercises for you.

Way more fun. You actually learn the codebase and you stop being a spectator to your own project. #ClaudeCode

[Ashutosh Srivastava @h4shkat]

Realtime visualizer for @a1zhang’s RLMs is live!

Watch recursive decomposition, sub-LLM calls, and REPL state unfold in real time.

Built directly on the inference engine with interactive visualizations.

Thread with details 👇

[Amit Sharma @amt_shrma]

🚀 Hiring Alert: Postdoc Positions at Microsoft Research India!

I'm especially looking for candidates interested in trustworthy AI reasoning, agentic systems and verification.

If you hold a PhD in CS or a related field and want to work on research that truly matters, consider applying.

⏳ Applications are reviewed on a rolling basis through April 15

Happy to answer any questions through replies here.

[Abhinav Java @abhinav_java]

Legend says Fede Valverde is still running.

#REALMADRİD #halamadrid #ucl

[Bhuvan Sachdeva @SachdevaBhuvan]

Full paper accepted to CVPR!!😀

[Nalin Wadhwa @ ICLR 2026 @nalin_wadhwa]

🚨 Can we truly prove that LLMs won't leak your data or generate unsafe code?

Almost all LLM safety evals today only give estimates.
Sampling. Benchmarks. Leaderboards.
Zero guarantees ❌

We present BEAVER, the first system to provably bound LLM constraint satisfaction.

🧵👇

[Bhuvan Sachdeva @SachdevaBhuvan]

New paper: Understanding Task Transfer in Vision–Language Models
How does finetuning a model on one task affect its performance on other tasks?
@karan_uppal3 and @abhinav_java are presenting this work at Unireps, NeurIPS!!
📍 Ballroom 20D
⏰ 3:45 PM – 5:00 PM
Come and say Hi!🧵

[Abhinav Java @abhinav_java]

@SandraWachter5 Same goes for writing, painting, etc. I think humans will continue to do things that bring them joy regardless of the ease of arriving at the final outcome. Or at least I do strongly hope so.

[Abhinav Java @abhinav_java]

For once I was glad that a reviewer responded within two hours of posting a rebuttal 🥲

[Yongyuan Liang @cheryyun_l]

After 24 hours of complete silence except for that single social media statement, ICLR has now decided to disregard all author and reviewer discussions during the two week rebuttal period ���

Quite a surprising way to wrap up this year’s review process