Darshan Singh

Super excited to announce that our work got accepted to CVPR'25! Here is a teaser for now, stay tuned for a detailed thread 😉

Can LLMs Self-Verify? Much better than you'd expect. LLMs are increasingly used as parallel reasoners, sampling many solutions at once. Choosing the right answer is the real bottleneck. We show that pairwise self-verification is a powerful primitive. Introducing V1, a framework that unifies generation and self-verification: 💡 Pairwise self-verification beats pointwise scoring, improving test-time scaling 💡 V1-Infer: Efficient tournament-style ranking that improves self-verification 💡 V1-PairRL: RL training where generation and verification co-evolve for developing better self-verifiers 🧵👇

Our paper was desk rejected @NeurIPSConf! Even before the main deadline! "Non-academic title and abstract" 🙈 Thankfully, @SIGGRAPHAsia was around the corner and a perfect fit for our work on improving robustness of multi-subject multi-attribute layout-guided T2I models! 🧵1/9

Researchers consider themselves very successful if they win one test-of-time award (and one is more than enough). Ross @inkynumbers has been winning them nonstop over the past year: CVPR 2024, ICCV 2025, and now NeurIPS 2025, because winning just one was too easy for him! Having known him for many years (first as a climbing partner and then as a colleague), I can’t say I’m surprised. When he sets his mind to something, he perfects it, whether it is making the best vision model, climbing a 5.12d, or continuing the sally-up sally-down push-up challenge until the rest of the team gives up. And to all his collaborators who only worked with him remotely and didn’t get to see him in person every day: you missed out. He is fun to work with but he is even more fun in person. I'm attaching the proof below. I have some true gem videos of his goofy side that I won’t share (saving them for when I need to blackmail him), but here is a photo of Ross pretending to be a lizard under our office sun lamp. Congratulations to Ross and all his co-authors. #NeurIPS2025

The NeurIPS Test of Time Award recognizes papers that have made lasting contributions to machine learning. For 2025 the award goes to…(buff.ly/W4ohNLA) #NeurIPS2025 #NeurIPSanDiego"

Late life update 🚀 I started my PhD at @UCBerkeley after an incredible time at @GoogleDeepMind. It was exciting to work on Gemini over the past couple of years. These days I am interested in reasoning/improving RL, agents, and diffusion language models. Looking forward to contributing to open science. Also thrilled to be back in the Bay Area. Grateful to mentors, collaborators, and folks who supported me, @partha_p_t @PengchuanZ @nitish_gup @trevorcohn @xiangrenNLP @divy93t @ManishGuptaMG1, Parag Singla, friends, and family. Excited to be at @NeurIPSConf #NeurIPS2025 this week. Looking forward to meeting folks. Feel free to DM if you'd like to chat!

@NithishKannen Wow. This is super cool!!

At 15, it was my dream to become a professional athlete. Nano Banana Pro 🍌 brought it to life - the character consistency and finer details are just insane!! "generate a visual story of this person <insert image> becoming an Olympic 100m champion"

Really proud of what we have achieved with Gemini 3 🚀! The Gemini MM team has worked relentlessly across image 🖼️ and video 🎥 from pre-training to post-training to simply deliver the best multimodal in the world 👏! Looking forward to what you will build🫡!

The calm after the storm!

the calm before the storm @googledeepmind hq ✨✨✨

This is Gemini 3: our most intelligent model that helps you learn, build and plan anything. It comes with state-of-the-art reasoning capabilities, world-leading multimodal understanding, and enables new agentic coding experiences. 🧵

imagine having a sunday this good

Happy that @gaur_manu and @thought2vec have this opportunity for our work on revisiting self-retrieval for fine-grained image captioning! Thread 🧵: x.com/gaur_manu/stat… arXiv: arxiv.org/abs/2409.03025 Project page: katha-ai.github.io/projects/no-de…

The goal of any good theory or new knowledge is to reduce uncertainty or entropy of a field. Given how many papers published each year related to machine/artificial intelligence and how many different opinions about this subject, I hope our new book is a denoiser, not a diffuser.

Excited to share I have joined CMU’s Robotics Institute as a Master’s student! The past few months were tough with visa uncertainties, but I’m happy to be finally here 😇 I look forward to working on native visual reasoning + other cool RL and VLM stuff 🚀 If you’re around campus, let’s hang out!

👀

Still can’t believe it. Our work on uncovering plagiarism in AI generated research received the outstanding paper award at ACL!! Effort led by, and envisioned by the amazing @tarungupta360!

A large contingent from IIITH’s Computer Vision Lab participated at the Conference on Vision and Pattern Recognition (CVPR) last month in Nashville. Read on about the cutting edge research that was presented and why it’s a big deal in the vision circles. blogs.iiit.ac.in/cvpr-2025/

Checkout this amazing work by my colleagues @Harman26Singh, @Pragya2k et al, on building robust LLM reward models by disentangling true attributes from spurious ones!!

Can text-to-image Diffusion models handle surreal compositions beyond their training distribution? 🚨 Introducing ComposeAnything — Composite object priors for diffusion models 📸 More faithful, controllable generations — no retraining required. 🔗zeeshank95.github.io/composeanythin… 1/9

🔔New @CVPR paper evaluating compositional reasoning of Video-LLMs on 10s, action-packed clips! 🥁 VELOCITI features 7 tests to disentangle and assess the comprehension of people, actions, and their associations across multiple events. katha-ai.github.io/projects/veloc… 🧵 1/9 #CVPR2025