Harsha Kokel's Avatar

Harsha Kokel

@kokel.bsky.social

AI and Sequential Decision Making @ IBM Research, San Jose. New mom.

160 Followers  |  144 Following  |  32 Posts  |  Joined: 12.11.2024  |  2.1239

Latest posts by kokel.bsky.social on Bluesky

Update: Submission deadline extended to November 7th!

πŸ”— Learn more and submit: plan-fm.github.io

31.10.2025 05:24 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

This is a great opportunity to showcase your research and engage with a vibrant community pushing the boundaries of intelligent systems.

πŸ”— Learn more and submit: plan-fm.github.io/2026

#AAAI2026 #PLANFM #FoundationModels

27.10.2025 18:27 β€” πŸ‘ 0    πŸ” 1    πŸ’¬ 0    πŸ“Œ 0

🧠 The aim of this bridge is to share resources and foster learning across communities.
πŸ“š It’s a non-archival venue β€” we welcome already published work, under-polished ideas, and papers currently under review.

#CallForPapers #CFP

27.10.2025 18:27 β€” πŸ‘ 0    πŸ” 1    πŸ’¬ 1    πŸ“Œ 0
Post image

πŸš€ Calling all researchers in planning and reasoning with foundation models!

If you're working on foundational models for planning and reasoning in natural language, we invite you to submit your work to PLAN-FM @aaai.org 2026

#AIPlanning #Reasoning

27.10.2025 18:27 β€” πŸ‘ 0    πŸ” 1    πŸ’¬ 1    πŸ“Œ 1
Preview
I’m an award-winning mathematician. Trump just cut my funding. The β€œMozart of Math” tried to stay out of politics. Then it came for his research.

I wrote an op-ed on the world-class STEM research ecosystem in the United States, and how this ecosystem is now under attack on multiple fronts by the current administration: newsletter.ofthebrave.org/p/im-an-awar...

18.08.2025 15:45 β€” πŸ‘ 793    πŸ” 324    πŸ’¬ 18    πŸ“Œ 32

Agree. Wondering if WFH parents have some hacks for context switching! Or do their neurons just adapt?

18.08.2025 05:34 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Super insightful. Thank you!

15.08.2025 21:20 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Poster announcing NYC RL day!

Poster announcing NYC RL day!

You know how RL is that whole big thing nowadays?

Present your work at the first-ever New York Reinforcement Learning Workshop (NYRL), co-organized by Amazon, Columbia Business School & NYU Tandon School of Engineering.

ny-rl.com!

25.07.2025 22:22 β€” πŸ‘ 26    πŸ” 4    πŸ’¬ 1    πŸ“Œ 0

Wonder if the estimation gap is this large for people who think AI does not help them in their work.

11.07.2025 22:23 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Take away #2 is 🀯

11.07.2025 22:22 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Preview
U.S. says it is now monitoring immigrants' social media for antisemitism Effective immediately, the government says it will begin screening immigrant social media for activity that officials think indicates support for antisemitism.

Effective immediately, the government says it will begin screening immigrant social media for activity that officials think indicates support for antisemitism.

09.04.2025 15:10 β€” πŸ‘ 8069    πŸ” 3978    πŸ’¬ 3435    πŸ“Œ 2376

πŸŽ™οΈ Title: Reasoning for Multimodal Semi-structured Tables
πŸ‘¨β€πŸ« Speaker: Dr. Vivek Gupta, Assistant Professor of Computer Science, Arizona State University (ASU).
πŸ“… Time: Tuesday, April 1st, 11:00 AM - 12:00 PM EDT, Online event

01.04.2025 03:24 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image

πŸš€ Tabular Data, Semantics, and AI seminar Alert πŸš€

Join us tomorrow to explore cutting-edge techniques for reasoning with multimodal semi-structured tables, a key challenge in making sense of complex, real-world data.

πŸ”— Link to attend: ibm.biz/tabular_data...

01.04.2025 03:23 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
Post image

Happy Chetichand, Hindu new year, Chaitra Navratri and all the festivals named in this image below.

31.03.2025 03:38 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Do you think it would help if the research community defined a new/different term for the intermediate steps; and not call it β€œreasoning” ? Curious.

20.03.2025 19:34 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
ACPBench: Reasoning about Action, Change, and Planning
YouTube video by Harsha Kokel ACPBench: Reasoning about Action, Change, and Planning

✍️ Read the blog: ibm.github.io/ACPBench/blo...
πŸ“œ Read the paper: arxiv.org/abs/2410.05669
▢️ Watch the talk: youtu.be/zlIOeYlo52M
πŸ€— Access the dataset: huggingface.co/datasets/ibm...

Joint work with Michael Katz, Kavitha Srinivas, and Shirin Sohrabi.

#AI #MachineLearning #LLMs

18.03.2025 23:35 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

πŸ” Our AAAI paper, ACPBench, introduces a benchmark designed to rigorously evaluate LLMs on core reasoning skills required for planning.
βœ… 7 tasks
βœ… 13 domains: includes IPC domains as well as one novel domain

#Planning #Reasoning

18.03.2025 23:35 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

πŸš€ ACPBench: A New Benchmark for Evaluating LLMs in Planning πŸš€

Large Language Models (LLMs) are increasingly being used as agents for workflow orchestration and decision-making in planning-heavy domains. But how well do they actually reason about action, change, and planning?

18.03.2025 23:35 β€” πŸ‘ 3    πŸ” 1    πŸ’¬ 1    πŸ“Œ 0

Congratulations. ☺️ Wish you a very health and rested pregnancy. You are such an inspiration; you are gonna be a great mom. ❀️

18.03.2025 04:17 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Preview
Bridging Planning and Reasoning in Natural Languages with Foundational Models

πŸ“… Mark your calendars and get ready to dive into the future of AI Planning! Find more information on our website: plan-fm.github.io

πŸš€ See you there!

22.02.2025 01:24 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Organizers:
Harsha Kokel, Shirin Sohrabi, Soham Dan, Manling Li, Yu Su

Advisors:
Biplav Srivastava, Sriraam Natarajan, Subbarao Kambhampati

22.02.2025 01:24 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

πŸ“š Tutorials to boost your knowledge in AI Planning, LLMs, and Language Agents.

🍽️ Exclusive for Students: Lunch-with-a-Mentor! Enjoy a unique opportunity to connect with AI mentors over lunch, ask questions, and get career insights.

22.02.2025 01:24 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

πŸ’‘ Plus, don’t miss our hour-long panel discussion on "Planning in Natural Language with LLMs"β€”where top minds will tackle the latest breakthroughs and challenges in AI Planning!

22.02.2025 01:24 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

Join us for an electrifying talk by Karen Myers on "AI Planning: From Shakey Foundations to a Chatty Future”

We also have invited talks from two incredible experts:
πŸ”₯ Swaroop Mishra
πŸ”₯ Mark "Mak” Roberts

22.02.2025 01:24 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
Post image

ttending AAAI next week?

πŸš€ Get Ready for an Epic AI Event! πŸ€–πŸŽ‰

PLAN-FM Bridge on Feb 26th.

We’ve got something truly special lined up for you!

#AAAI2025 #Planning #RL #PlanFM #LLMs #Reasoning

22.02.2025 01:24 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

Not sure if it is attained or maintained on purpose. But I feel same thing for uninformed support too; its worse than regular support.

29.12.2024 04:36 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
Post image

RLHF just took us by surprise!!

14.12.2024 20:25 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

🚨 Final CFP and Extension: PLAN-FM @ AAAI 🚨

Submit your relevant work on planning in natural language before the AI crowd heads to NeurIPS.

Submission Deadline: Dec 9th AoE.

#Planning
#AAAI
#NLP
#Reasoning
#LLM
#FoundationalModels
#AI

03.12.2024 22:31 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Overview of paper browser. A cluster for reinforcement learning is selected.

Overview of paper browser. A cluster for reinforcement learning is selected.

Paper Browser: only papers assigned to "physical models - physics" are shown.

Paper Browser: only papers assigned to "physical models - physics" are shown.

Paper Browser: Filtered by author "Hoover" and detail is shown

Paper Browser: Filtered by author "Hoover" and detail is shown

Paper Brower: ZOOOOM in

Paper Brower: ZOOOOM in

🎺 Here comes the official 2024 NeurIPS paper browser:
- browse all NeurIPS papers in a visual way
- select clusters of interest and get cluster summary
- ZOOOOM in
- filter by human assigned keywords
- filter by substring (authors, titles)

neurips2024.vizhub.ai

#neurips by IBM Research Cambridge

03.12.2024 17:01 β€” πŸ‘ 59    πŸ” 22    πŸ’¬ 5    πŸ“Œ 4

Abstract submission deadline: Sunday, December 1st, 2024
Paper submission deadline: Sunday, December 8th, 2024 (extended)
Paper acceptance notification: Monday, December 16th, 2024 (extended)

27.11.2024 20:52 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

@kokel is following 20 prominent accounts