Amit Agarwal

We are bringing together the brightest minds in CV, NLP, and IR to bridge the gap between "Multimodal Frontier Models/Tools", "Agents" and "Production-Scale Intelligence" at GRAIL-V workshop @CVPR 26 🔥 Keynotes from: 🌟 Kristen Grauman (@UTAustin ) 🌟 @mohitban47 (@unc_ai_group ) 🌟 @DanRothNLP (@_PennAI ) 🌟 @scottyih (@Meta @AIatMeta ) 🎤 Industry Panel: Moderated by @ravisujith (VP, @Oracle AI) If you resonate with these problems and are solving it, we would like to see your work ! Call for Papers is open. Submission - 5th March, 2025 Website - grailworkshops.github.io OpenReview - openreview.net/group?id=thecv… Organizing Team - @sarahookr @aliceoh @jyotika @Hitesh_LPatel Vivek Gupta, Vivek Srikumar, Tao Sheng #CVPR2026 #AI #ComputerVision #LLM #Agents #Multimodal #Research #MachineLearning #ICLR2026 #ICML2026 #VisionLanguage

#CVPR2026 is just around the corner! If you are heading to Denver, join us at GRAIL-V @CVPR on 3rd June, for a morning dedicated to solving the hard parts of Agentic Systems - bridging the gap between Computer Vision, NLP, and Information Retrieval We have a packed schedule featuring foundational keynotes and a deep-dive industry panel focused on moving from frontier research to production-scale agents for @CVPRConf 📅 Date: Wednesday, June 3, 2026 ⏰ Time: 7:30 AM – 12:30 PM 📍 Location: Room 506, Colorado Convention Center, Denver 🔗 Details & Full Schedule: lnkd.in/g4JaU5x6 🔥 Keynotes from: 🌟 Kristen Grauman (@UTAustin) 🌟@mohitban47 (@unc_ai_group) 🌟@DanRothNLP (@_PennAI ) 🌟@scottyih (@Meta @AIatMeta ) 🎤 Industry Panel: @ravisujith(GVP, @Oracle AI) , @krishnanvijay (@turingcom ), @Kenneth_Marino (@UUtah ), @MingHsuanYang (@ucmerced @GoogleDeepMind ) Organizing Team - @amitpinaki @sarahookr @aliceoh @jyotika @Hitesh_LPatel @keviv9 , @karandua Vivek Srikumar, Tao Sheng #CVPR2026 #AI #ComputerVision #LLM #Agents #Multimodal #Research #MachineLearning #ICLR2026 #ICML2026 #VisionLanguage

@sukh_saroy Thankuzzz

🚨 BREAKING: Someone just open sourced a tool that gives your AI agent a complete nervous system for your codebase and it's not a code search. It's called GitNexus and it's not a README explainer. It's a real knowledge graph engine that maps every dependency, call chain, execution flow, and breaking change risk in your entire codebase then feeds it directly into Claude Code, Cursor, and Windsurf via MCP. Here's what it actually does: → Indexes your entire repo into a knowledge graph in one command → Tells your AI agent exactly what breaks if you touch any function → Maps every upstream dependency, import, and call chain automatically → Traces full execution flows from entry points through the entire stack → Shows blast radius analysis with confidence scores before you ship → Works with 12 languages including TypeScript, Python, Go, Rust, and Java → Runs entirely locally - zero network calls, zero code uploaded anywhere Here's the wildest part: Your AI agent edits a function. It doesn't know 47 other functions depend on its return type. Breaking changes ship. GitNexus fixes this by precomputing all relationships at index time - so one tool call returns the complete picture instead of the agent running 10 queries and still missing something. Even smaller, cheaper models get full architectural clarity. You don't need GPT-5 when your tools are this good. You're using Cursor and Claude Code daily and shipping blind edits. GitNexus closes that gap. One command. Fully local. The nervous system your AI agent was always missing just got open sourced. 9,400+ GitHub stars. 1,200+ forks. Already trending. 100% Open Source. (Link in the comments)

Submission Deadline has been extended for GRAIL-V workshop at @CVPR 2026 Push through your latest research and findings for Multimodal Models and Agents for both archival and non-archival tracks ! Looking forward to see you at #CVPR2026 ! Link - grailworkshops.github.io #ComputerVision #AI #Multimodal #ICLR #ICML #Embeddings #Benchmarks

🚀 Call for Papers | SURGeLLM Workshop @aclmeeting 2026 #ACL2026 Structured data is everywhere yet still underrepresented in how we build and evaluate LLMs. SURGeLLM is a new workshop dedicated to making tables, time series, graphs, charts, maps, flowcharts, and other structured artifacts first-class citizens in modern LLM systems. If you’re working on structure-aware LLMs, this workshop is for you. 🔍 We welcome work on - Structured and symbolic reasoning, multimodal grounding, data-centric LLMs, semantic parsing, visual analytics, and evaluation methodologies for structured artifacts. 📅 Submission deadline: March 22 📝 Details & submission: surgellm.github.io/acl2026/cfp/ (via @openreviewnet ) 🎤 Invited Speakers: @DanRothNLP , @hengjinlp, @LisaAmini1 , @hamidpalangi 💼 Supported By: @SnorkelAI We’re excited to share that SURGeLLM is officially sponsored by Snorkel AI, supporting best paper awards and workshop activities. Thanks to Alexander Ratner, @paroma_varma, @fredsala, @_Incynthia, and @KobieWon 👥 Organizers @keviv9 , @kaize0409, @harsha_kokel, Yue Zhao, @amitpinaki, Yu Wang, Michael Glass, @yuz9yuz, Kavitha Srinivas, Xiusi Chen, @oktie, Qi Zhu, @ShuaichenChang, @yuanhypnosluo Let’s push the boundaries of structure-aware, trustworthy, and multimodal LLMs - together.💡 #NLPProc #ACL2026 @aclmeeting , @ReviewAcl,

Deadline extended! Submit your work at Grail-V to be at @CVPR. #CVPR2026 @amitpinaki @Hitesh_LPatel

🚨 5 Days to submit to the GRAIL-V Workshop at @CVPR' 26 ! We are calling for research (Archival & Non-Archival) on Multimodal Retrieval, Generation, Planning , Reasoning and Agentic Systems/Tools, along with research one benchmarks/evaluation. Join our amazing lineup: KristenGrauman (@UTAustin ), @mohitbansal (@UNC @unc_ai_group ), @danroth (@Penn @_PennAI ), @scottyih (@AIatMeta ) & @ravisujith (@Oracle ) . 📅 Deadline: March 5 🔗 Submit: openreview.net/group?id=thecv… Read our vision here: blogs.oracle.com/ai-and-datasci… #AIResearch #ComputerVision #GenAI #CVPR2026 #Agents #Multimodal #Workshop

@CVPR @Coding_Black GRAIL-V workshop is accepting submissions on Multimodal Models and Agentic Vision systems- #submit" target="_blank" rel="nofollow noopener">grailworkshops.github.io/cfp/#submit

@Coding_Black If paper was rejected, then ACs suggested findings workshop. If paper accepted, ignore findings recommendation. Hope this helps.

#CVPR2026 final decisions are out! Available for now only via email. Good luck🤞

@JyotikaSingh_ @dr_chizhang - sharing incase your labs and students find it relevant

@JyotikaSingh_ i got the wrong one initially, but here you go

We are bringing together the brightest minds in CV, NLP, and IR to bridge the gap between "Multimodal Frontier Models/Tools", "Agents" and "Production-Scale Intelligence" at GRAIL-V workshop @CVPR 26 🔥 Keynotes from: 🌟 Kristen Grauman (@UTAustin ) 🌟 @mohitban47 (@unc_ai_group ) 🌟 @DanRothNLP (@_PennAI ) 🌟 @scottyih (@Meta @AIatMeta ) 🎤 Industry Panel: Moderated by @ravisujith (VP, @Oracle AI) If you resonate with these problems and are solving it, we would like to see your work ! Call for Papers is open. Submission - 5th March, 2025 Website - grailworkshops.github.io OpenReview - openreview.net/group?id=thecv… Organizing Team - @sarahookr @aliceoh @jyotika @Hitesh_LPatel Vivek Gupta, Vivek Srikumar, Tao Sheng #CVPR2026 #AI #ComputerVision #LLM #Agents #Multimodal #Research #MachineLearning #ICLR2026 #ICML2026 #VisionLanguage

With #CVPR2026 results and decisions released, you still have time to participate in the conference and showcase your work. GRAIL-V works invites state-of-the-art multimodal LLMs, Agents, Generations Models that work on various aspects of retrieval, reasoning, creativity and much more. Submission Deadline - 5th March Submission Link - openreview.net/group?id=thecv… Workshop Website - grailworkshops.github.io #Multimodal #ComputerVision #Agents #ICLR #ICML #ACL #EMNLP #ECCV #ICCV

This transition—from Passive Observation to Active Verification—is exactly what we are exploring at the GRAIL-V Workshop @CVPR 26. If you are working on: 1. Tool-Use for Vision 2. Visual Grounding 3. Verifiable Agents 4. We want your papers! 🗓️ Deadline: March 5 🔗 grailworkshops.github.io #CVPR2026 #Gemini #GenAI #CV #ICLR2026

🚨 "Passive Vision" is officially outdated. @GoogleResearch /@Google /@GoogleDeepMind just launched Agentic Vision in Gemini 3 Flash. The model no longer just "looks" at an image. It investigates it. • It writes code to zoom in. • It draws bounding boxes to count. • It verifies its own hallucinations. Here is why this is a big deal 🧵 #CVPR2026 #computervision #agents

Standard VLMs are "One-Shot Guessers." Agentic Vision is a "Multi-Step Verifier." By using a Think → Act → Observe loop, the model can "ground" its answer in actual pixels rather than statistical probability. This is the only way to solve the "Hallucination" problem in critical tasks.

@karpathy This is so true and hard to know where its heading for engineers

A few random notes from claude coding quite a bit last few weeks. Coding workflow. Given the latest lift in LLM coding capability, like many others I rapidly went from about 80% manual+autocomplete coding and 20% agents in November to 80% agent coding and 20% edits+touchups in December. i.e. I really am mostly programming in English now, a bit sheepishly telling the LLM what code to write... in words. It hurts the ego a bit but the power to operate over software in large "code actions" is just too net useful, especially once you adapt to it, configure it, learn to use it, and wrap your head around what it can and cannot do. This is easily the biggest change to my basic coding workflow in ~2 decades of programming and it happened over the course of a few weeks. I'd expect something similar to be happening to well into double digit percent of engineers out there, while the awareness of it in the general population feels well into low single digit percent. IDEs/agent swarms/fallability. Both the "no need for IDE anymore" hype and the "agent swarm" hype is imo too much for right now. The models definitely still make mistakes and if you have any code you actually care about I would watch them like a hawk, in a nice large IDE on the side. The mistakes have changed a lot - they are not simple syntax errors anymore, they are subtle conceptual errors that a slightly sloppy, hasty junior dev might do. The most common category is that the models make wrong assumptions on your behalf and just run along with them without checking. They also don't manage their confusion, they don't seek clarifications, they don't surface inconsistencies, they don't present tradeoffs, they don't push back when they should, and they are still a little too sycophantic. Things get better in plan mode, but there is some need for a lightweight inline plan mode. They also really like to overcomplicate code and APIs, they bloat abstractions, they don't clean up dead code after themselves, etc. They will implement an inefficient, bloated, brittle construction over 1000 lines of code and it's up to you to be like "umm couldn't you just do this instead?" and they will be like "of course!" and immediately cut it down to 100 lines. They still sometimes change/remove comments and code they don't like or don't sufficiently understand as side effects, even if it is orthogonal to the task at hand. All of this happens despite a few simple attempts to fix it via instructions in CLAUDE . md. Despite all these issues, it is still a net huge improvement and it's very difficult to imagine going back to manual coding. TLDR everyone has their developing flow, my current is a small few CC sessions on the left in ghostty windows/tabs and an IDE on the right for viewing the code + manual edits. Tenacity. It's so interesting to watch an agent relentlessly work at something. They never get tired, they never get demoralized, they just keep going and trying things where a person would have given up long ago to fight another day. It's a "feel the AGI" moment to watch it struggle with something for a long time just to come out victorious 30 minutes later. You realize that stamina is a core bottleneck to work and that with LLMs in hand it has been dramatically increased. Speedups. It's not clear how to measure the "speedup" of LLM assistance. Certainly I feel net way faster at what I was going to do, but the main effect is that I do a lot more than I was going to do because 1) I can code up all kinds of things that just wouldn't have been worth coding before and 2) I can approach code that I couldn't work on before because of knowledge/skill issue. So certainly it's speedup, but it's possibly a lot more an expansion. Leverage. LLMs are exceptionally good at looping until they meet specific goals and this is where most of the "feel the AGI" magic is to be found. Don't tell it what to do, give it success criteria and watch it go. Get it to write tests first and then pass them. Put it in the loop with a browser MCP. Write the naive algorithm that is very likely correct first, then ask it to optimize it while preserving correctness. Change your approach from imperative to declarative to get the agents looping longer and gain leverage. Fun. I didn't anticipate that with agents programming feels *more* fun because a lot of the fill in the blanks drudgery is removed and what remains is the creative part. I also feel less blocked/stuck (which is not fun) and I experience a lot more courage because there's almost always a way to work hand in hand with it to make some positive progress. I have seen the opposite sentiment from other people too; LLM coding will split up engineers based on those who primarily liked coding and those who primarily liked building. Atrophy. I've already noticed that I am slowly starting to atrophy my ability to write code manually. Generation (writing code) and discrimination (reading code) are different capabilities in the brain. Largely due to all the little mostly syntactic details involved in programming, you can review code just fine even if you struggle to write it. Slopacolypse. I am bracing for 2026 as the year of the slopacolypse across all of github, substack, arxiv, X/instagram, and generally all digital media. We're also going to see a lot more AI hype productivity theater (is that even possible?), on the side of actual, real improvements. Questions. A few of the questions on my mind: - What happens to the "10X engineer" - the ratio of productivity between the mean and the max engineer? It's quite possible that this grows *a lot*. - Armed with LLMs, do generalists increasingly outperform specialists? LLMs are a lot better at fill in the blanks (the micro) than grand strategy (the macro). - What does LLM coding feel like in the future? Is it like playing StarCraft? Playing Factorio? Playing music? - How much of society is bottlenecked by digital knowledge work? TLDR Where does this leave us? LLM agent capabilities (Claude & Codex especially) have crossed some kind of threshold of coherence around December 2025 and caused a phase shift in software engineering and closely related. The intelligence part suddenly feels quite a bit ahead of all the rest of it - integrations (tools, knowledge), the necessity for new organizational workflows, processes, diffusion more generally. 2026 is going to be a high energy year as the industry metabolizes the new capability.

@EXM7777 @Hitesh_LPatel

turn this thread into instructions for your claude[.]md file this might just change your life

This is a perfect example of how Agents can be made to reliably help and improve with proper grounding! @GoogleResearch AgenticVision launch affirms we need more to be done in this space as it is just starting to boom. If you are solving similar problems- you should consider submitting at GRAIL-V workshop @CVPR 2026! We are bring researcher and practitioners to push the boundary for Agentic Vision Website - grailworkshops.github.io/cfp/ Submission Deadline - 5th March #AgenticVision #AgenticAI #CVPR2026 #ICLR2026 #ICML2026

@ceekz I think you would want to checkout our workshop as we exactly want to address this problem where deep research agents cannot be trusted grailworkshops.github.io

NLP関係者の98%が衝撃を受ける論文が公開された👀 [2601.18724] HalluCitation Matters: Revealing the Impact of Hallucinated References with 300 Hallucinated Papers in ACL Conferences arxiv.org/abs/2601.18724

@cheryyun_l Great to see this work and benchmark in cross-modal reasoning! We are hosting GRAIL-V workshop at CVPR 2026 addressing the same problems grailworkshops.github.io You and your team must explore to share more of your findings and research to get the community think on this !

Unified multimodal models can generate text and images, but can they truly reason across modalities? 🎨 Introducing ROVER, the first benchmark that evaluates reciprocal cross-modal reasoning in unified models, the next frontier of omnimodal intelligence. 🌐 Project: roverbench.github.io 📄 Paper: arxiv.org/abs/2511.01163 📂 Benchmark: huggingface.co/datasets/chery…