wing.nus

546 posts

wing.nus banner
wing.nus

wing.nus

@wing_nus

Web IR / NLP Group at the National University of Singapore

Singapore Katılım Temmuz 2012
301 Takip Edilen605 Takipçiler
wing.nus
wing.nus@wing_nus·
🔍 Key insight: In AI systems, the bottleneck is no longer generation but oversight. As AI autonomy grows, validation, monitoring, and accountability burdens increase. Mitigation requires bounded autonomy, calibrated reliance, governance, and AI literacy. 🧵4/5
English
1
0
0
14
wing.nus
wing.nus@wing_nus·
We were delighted to host Philipp Mayr (GESIS – Leibniz Institute for the Social Sciences) at our group meeting! 🎉 He presented the Knowledge Technologies department, the Information & Data Retrieval team, the OMINO project, and two Scholarly Document Processing projects. 🧵1/5
wing.nus tweet media
English
1
0
7
142
wing.nus
wing.nus@wing_nus·
Bottom line: Better fact-checking isn’t about more web access. It’s about finding what’s missing. Search the gaps → better Notes. If you’re working on misinformation, moderation, or Community Notes, we’d love your thoughts. (6/6)
English
0
0
0
36
wing.nus
wing.nus@wing_nus·
Human evaluation (N=100): 🏆 69% win rate vs human-written helpful notes 📈 Helpfulness: 3.87 vs 3.36 📌 Biggest gain: better context Also, 59% win vs generic web agents. Search strategy matters. (5/6)
wing.nus tweet media
English
1
0
0
45
wing.nus
wing.nus@wing_nus·
🚨 Can smarter search beat generic web-search LLMs for Community Notes? Same model. Same web access. Different search strategy. We built GitSearch to test this. 📄 arxiv.org/abs/2602.08945 (1/6)
wing.nus tweet media
English
1
1
5
678
wing.nus
wing.nus@wing_nus·
📌 Final thought Trust is a learnable, usable signal for LLM agents. Towards reliable multi-agent systems, we must teach agents who to believe — not just how to reason. 🔗 arXiv: arxiv.org/abs/2601.21742 🏠 Github: github.com/skyriver-2000/… 💬 Thoughts welcome! 🧵6/n
English
0
0
1
58
wing.nus
wing.nus@wing_nus·
⚠️ Why this matters Without trust modeling: ❌ agents collapse under social pressure ❌ confident hallucinations dominate ❌ adversarial peers win With ECL: ✅ agents resist blind conformity ✅ trust becomes an explicit reasoning signal Analytic experiments verify this 🧵5/n
wing.nus tweet media
English
1
0
1
67
wing.nus
wing.nus@wing_nus·
🔥 Thrilled to announce ECL, a framework for LLMs to reason with trust in multi-agent systems 📖 Key Takeaways •We introduce interaction history for LLMs to judge peer reliability and selectively refer to them •We decouple trust estimation and conditioned decision-making 🧵1/n
wing.nus tweet media
English
2
0
7
534
wing.nus
wing.nus@wing_nus·
🔮Meet IzzyViz, a new way to see how transformer models think. 🔥Attention heatmaps | 🧠Head & layer comparison | 📊Run-to-run stability | ⏳Training-time evolution | 🎯Key-region detection 🔗Repo: github.com/WING-NUS/IzzyV…
wing.nus tweet media
English
0
0
2
144
wing.nus
wing.nus@wing_nus·
📢 Excited to share our accepted EMNLP 2025 papers from the NUS WING group! 🎉 See you in Suzhou! #EMNLP2025
wing.nus tweet media
English
0
0
7
1.1K
wing.nus
wing.nus@wing_nus·
High-quality knowledge can be "distilled"! We used GPT-4o to generate a knowledge base for a smaller Llama3.1-8B. This "distillation" significantly boosted its performance, enabling efficient, high-quality narration. 🧵[4/n]
wing.nus tweet media
English
1
0
0
99
wing.nus
wing.nus@wing_nus·
Thrilled to share that our paper, "KAHAN: Knowledge-Augmented Hierarchical Analysis and Narration," is accepted at #EMNLP2025 Findings! In this work, we built a framework that uses LLMs as domain experts to hierarchically extract insights from tables. 🧵[1/n]
English
1
0
2
144