Shivanand 🇮🇳

भारतीय || ಕನ್ನಡಿಗ || RTs/likes/Links are not endorsement.

[Shivanand 🇮🇳 @ssheshap]

#CVPR2027 submissions crossed 40,000?

[Hyper.AI @HyperAI_News]

14 years of CVPR submissions

2013 → 2026:
Submissions: 1,798 → 16,092 (9x)
Acceptance rate: 26.3% → 25.4% (peaked at 29.9% in 2014–2017)

Bold call: will CVPR 2027 submissions break 20,000?

#CVPR #CVPR2027 #ComputerVision #AIResearch #AcademicChatter

[🚩Mohan Gowda🇮🇳 @Mohan_HJS]

Why Kannada organizations are silent?

[Shivanand 🇮🇳 @ssheshap]

@SriKrishna1436 If at all #Yash can do 16 movies in a single year.

[Shivanand 🇮🇳 @ssheshap]

@y_m_asano @iclr_conf Also, 2/3 of 5th Nov is over (EST).

[Tianqin Li @ CMU @Jack_Litq]

As we are approaching ICLR review release date, I’m very curious what if the paper / conference push their submission deadlines closer to the actual conference date so that papers presented at the conference can stay up to date.

For example, ICLR 27 will be held in Apr 2027, why not setting the submission deadlines as Jan 15, 2027, or even February 2027? So a three / four month old paper can still reflect somewhat state-of-the-art results, whereas an 8 month old paper is most likely outdated. That’s why sometimes many people enjoy attending workshops as they are submitted much more recently.

It’s great to see ICLR this year push the final decision before the new year! That’s quite convenient for people applying for graduate schools :) . But I really hope the next tier one conference papers can be more up-to-date …

(Or perhaps let’s just use X~ 🤣

#ICLR27 #AI #Research

[Shivanand 🇮🇳 @ssheshap]

@priyankac19 Education ministers # 1, 3, 5, 9 (approx. 16/80 yrs).

https://t.co/s5rhjnbLsv

[Shivanand 🇮🇳 @ssheshap]

@sunandavashisht PoK

[Kate Saenko @kate_saenko_]

My group at FAIR (Meta) is looking for a summer research intern. If you have experience with SAM/object segmentation, especially in microscopy images or similar domains, and are excited to contribute to the next generation of SAM models, apply here: https://t.co/dMOQ45YOmI

[Rikiya Takehi @rikiyatakehi]

We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.

We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: https://t.co/bOWCcFhCVq

[Kevin Patrick Murphy @sirbayes]

This is a great post! A must read!
https://t.co/SmN3oJbW68

[Michael C. Frank @mcxfrank]

New blogpost: my lab's AI policies and norms.

I'm trying to figure out how researchers should make use of AI to accelerate discovery without letting it replace human understanding or falling off the "slop cliff." Here are some ideas.

https://t.co/etz0DQvnHL

[Jacob X. Li @jacobli99]

someone I really admire said researchers should share more of the messy day-to-day instead of only the polished results

so in the spirit of open science, nothing worked today.

[Pedro Domingos @pmddomingos]

Writing is thinking. If you outsource your writing, you outsource your thinking. (@jaschasd)

[` @chixxsays]

This edit of PM Modi is an absolute banger. The editor deserves an Oscar for this one

25 YEARS SEVA STREAK

[Aditya Raj Kaul @AdityaRajKaul]

#BREAKING: India’s Women Hockey Team win Gold beating defending champions China 1-0 in the Asian Games Final. India qualify for the 2028 LA Olympics. India’s massive golden glory. Hockey Gold for Indian Women Team after 44 years. Congratulations to the stars! 🇮🇳

[ಹಂಸಾನಂದಿ Hamsanandi हंसानन्दि @hamsanandi]

D. Narrative: "Kannadigas are shiva worshippers , Rama Krishna Vishnu came in much later, post 10th C. They are north Indian gods"

Truth: Two Kadamba inscription from around the same time ~450 CE

Halmidi inscription in Kannada - Praises Vishnu in the mangala padya
Talagunda inscription in Samskrta - Praises Shiva in the Mangala padya

Before falling prey to D-virus, know your facts right!

#Halmidi #Talagunda #Kadamba #inscription #ಕದಂಬ #ಶಾಸನ #ತಾಳಗುಂದ #ಹಲ್ಮಿಡಿ

[Shivanand 🇮🇳 @ssheshap]

@ajeetbharti Justice + Politics = Justics

Ajeet Bhai, please understand the emotions. Support them.

[Tech with Mak @techNmak]

CMU is putting its new AI Agents course on YouTube for free.

It is 11-768, taught this Fall by Graham Neubig and Daniel Fried, and this is one of the first agent courses I’ve seen that goes all the way from building the harness to training the model behind it.

The early part covers the things you would expect: tool use, context management, skills, memory and planning.

Then it gets much more interesting.

There are separate sections on coding agents, GUI agents and deep-research agents, followed by supervised fine-tuning, reinforcement learning, RL systems, sandboxing, credential management, OpenHands, LangGraph, observability, multi-agent interaction, human-agent interaction and search.

The assignments are probably the best description of the course.

First, build an agent harness.

Then build evaluations for it.

Then train an agent with reinforcement learning.

The first assignment is already public on GitHub, including the starter code. The lectures are being uploaded to YouTube as the semester runs, and Graham Neubig has said the full set of videos will go into the public playlist.

This is considerably more useful than learning one agent framework and assuming you understand agents.

You get to see the whole problem: the model, the harness around it, the tools and memory it uses, the evals that tell you whether it works, and the training that can change the model itself.

Course:
https://t.co/dOk3OintCS

YouTube:
https://t.co/sfFYWv4S2E

[Ajeet Bharti @ajeetbharti]

That's precisely why it is the best among IITs. CJI Suryakant remarked that reservation for teachers was against the fundamental right to good education of the students.

[Jonathan Aldrich @JAldrichPL]

I am proud to say that I am exercising self-restraint and only submitting one ICLR paper this year