LCS2 Lab

Lab. for Computational Social Systems, a group led by @Tanmoy_Chak working on #SocialComputing #GraphMining & #NLProc

[LCS2 Lab @lcs2lab]

Excited to share that our work, #MonteCLoRA, has officially been merged into the #HuggingFace PEFT library! 🥳 https://t.co/1411PsNGwc

Build #peft from source to use it right away! 🚀

📜 Paper: https://t.co/vPAvMvYsrb
🤗 Docs: https://t.co/fHtm0NW4BM

[LCS2 Lab @lcs2lab]

🇧🇷 #LCS2 goes to #Rio 🇧🇷 Presenting our paper where we move beyond memoryless personalization → modeling user preferences as action-conditioned geometric walks with memory for better, user-aligned summaries. See you at #Riocentro 🚀 #Personalization #RepresentationLearning

[LCS2 Lab @lcs2lab]

Happy to announce that our paper has been accepted to #ICLR2026! 🎉
📜 Beyond Markovian Drifts: Action-Biased Geometric Walks with Memory for Personalized Summarization
👥 Parthiv Chatterjee, Asish Batha, Tashvi Patel, @sourish_rygbee, @Tanmoy_Chak
Congratulations to all authors!

[LCS2 Lab @lcs2lab]

If you're interested in multimodal reasoning, math LLMs, or vision-language models, do check out our paper! #MultimodalAI #LLMs #MathReasoning #VisionLanguage #AIForScience #VisualReasoning #SymbolicReasoning #GeometricReasoning

[LCS2 Lab @lcs2lab]

🚀 #EACL2026 Findings Sneak Peek 🚀
We're excited to present our latest work in the #EACL2026 Findings!
📜 SpatialMath: Spatial Comprehension-Infused Symbolic Reasoning for Mathematical Problem-Solving
👥 Ashutosh Bajpai, Akshat Bhandari, Akshay Nambi, @Tanmoy_Chak

[LCS2 Lab @lcs2lab]

📖 Paper: https://t.co/AJKJ8G1Fp6
✨Current multimodal LLMs don’t truly see math diagrams -- they approximate. We introduce SpatialMath, which bridges perception and reasoning, enabling more reliable visual mathematical problem-solving.

[LCS2 Lab @lcs2lab]

MLRBench aims to push the field of multilingual long-context reasoning toward more realistic evaluation. If you're in Morocco for #EACL2026 -- come say hi!
#LLMs #MultilingualAI #MultilingualNLP #LongContext #NLProc #EfficientAI #ScalableAI #RAG #Reasoning #EvaluationBenchmarks

[LCS2 Lab @lcs2lab]

🚀 #EACL2026 Sneak Peak Alert 🚀
We're excited to share a paper that we are presenting at #EACL2026 in #Morocco!
📜 Can LLMs Reason over Extended Multilingual Contexts? Towards Long-Context Evaluation Beyond Retrieval over Haystacks
👥 @AmeyHengle @prasNLP Soham Dan @Tanmoy_Chak

[LCS2 Lab @lcs2lab]

📊 Key Findings:
- Reasoning fails before retrieval
- Multilingual gap is real
- Effective context <30%: LLMs utilize only a fraction of their claimed context window
- RAG helps, but not enough
- Uncertainty reasoning is hardest

[LCS2 Lab @lcs2lab]

📢 New Media Feature 📢 We are pleased to share that Prof. Tanmoy Chakraborty @Tanmoy_Chak was recently invited to Insight, a program on @sansad_tv, to speak on the growing challenges posed by #AIGenerated synthetic content.
#ResponsibleAI #PublicPolicy

https://t.co/ckn8kjp6XM

[LCS2 Lab @lcs2lab]

🚀 New Podcast Alert 🎙️
Prof. Tanmoy Chakraborty @Tanmoy_Chak joins Rudraditya on The Inner Circle With Rudraditya podcast for a sharp, no-hype conversation on the realities, risks, and future of AI. #AI #MachineLearning #SLM #AIGovernance #NLProc
https://t.co/KSUw5bYoge

[LCS2 Lab @lcs2lab]

✨ Huge thanks to all collaborators and participants whose lived experiences shaped this work. We hope this contributes toward accessible, scalable, and culturally resonant digital mental health solutions, especially in low-resource settings.

[LCS2 Lab @lcs2lab]

🎉 New Paper Alert 🎉 We're excited to share that our paper "Here for You: A Co-Designed Mental Health Screening App for Indian University Students " has been accepted for publication in #JMIR #FormativeResearch! 🌟 @Tanmoy_Chak
https://t.co/dHmRGDTdsJ

[LCS2 Lab @lcs2lab]

🔍 Key Highlights (contd.)
- Ethical and regulatory alignment, including adherence to DPDP guidelines
- Demonstrates the power of the “nothing about us without us” principle in digital mental health

[LCS2 Lab @lcs2lab]

Happy to announce that our paper has been accepted to #ICLR2026! 🎉
📜 Beyond Markovian Drifts: Action-Biased Geometric Walks with Memory for Personalized Summarization
👥 Parthiv Chatterjee, Asish Batha, Tashvi Patel, @sourish_rygbee, @Tanmoy_Chak
Congratulations to all authors!

[LCS2 Lab @lcs2lab]

🎤 Keynote | Singapore Symposium on NLP #SSNLP @ #AAAI2026
🗨️ Enough of Scaling Laws! Let's Focus on Downscaling
👤 @Tanmoy_Chak
🔍 A call to rethink scale focusing on efficient, interpretable, and adaptable #SmallLanguageModels that rival large models without prohibitive costs.

[LCS2 Lab @lcs2lab]

#LCS2 is ready for #AAAI2026 🇸🇬 🦁
We are excited to be in Singapore this week to share our latest work across quantum-inspired NLP, brain-aligned AI, and efficient language models. If you're around #AAAI2026 or #SSNLP2026, drop by and say hello! #NLProc #AIResearch @Tanmoy_Chak

[LCS2 Lab @lcs2lab]

🎓 Tutorial | #AAAI2026
🧑‍💻 Brain-Inspired AI 2.0: Aligning Language Models Across Languages and Modalities
👥 Subba Reddy Oota, @Tanmoy_Chak, @ManishGuptaMG1, Raju S. Bapi
🧠 Exploring the intersection of computational cognitive neuroscience and AI. #BrainInspiredAI #CognitiveAI