Victor Xing (@VictorTXing) profile picture

Applied research @illuintech

Joined May 2018
Photo by Victor Xing (@VictorTXing)Photo by Victor Xing (@VictorTXing)Photo by Victor Xing (@VictorTXing)
There's been a lot of debate lately about whether you can tell AI-written literature from human writing @antoine_edy and I made a daily game to find out how good you are at it Try it out at fauxlio.app
🧩 Can you tell a real book from AI writing? We built a fun little game called fauxlio.app Two excerpts: one's a classic, one's AI, and your goal is to pick the real one. 5 new rounds a day, EN/FR, free, no sign-up. Tell @VictorTXing and me what you think! 🫶
1
1
4
102
Victor Xing retweeted
We introduce DAEDALUS! A new method that lets agents learn on tasks they invent themselves: no training tasks, no ground truth, no oracle. Find links to the paper, blog post, and more below ⬇️
1
2
6
111
Victor Xing retweeted
Our latest work at @illuintech is out! 🕺 DAEDALUS is a training-free pipeline that generates priors on any new environment, that an agent can use in-context at test time. No tasks/verifier/human-written guidelines required! Details and fav take-aways in the thread below 👇 (1/n)
1
3
11
362
Super excited to release Daedalus, our work on bootstrapping agent memory from self-generated tasks! Building agent memory needs practice tasks, but most environments have none. So we let the agent create its own. More in Antoine's 🧵 and our blog post illuin-tech.github.io/daedal…
🏛️ LLM agents in a new environment repeat the same mistakes. Meet DAEDALUS: agent memory bootstrapped from tasks the agent writes itself, with no training tasks or human-written guidelines. It makes 7 agents from various model families succeed more often, in fewer turns 🧵⬇️
2
2
8
391
I'm attending #ICLR2026 in Rio 🇧🇷 this week! I will be presenting our paper Experiential Reflective Learning for Self-Improving LLM Agents 🗓️ Monday Apr 27 · 2:50–3:35pm 📍 MemAgents, Room 205 arxiv.org/abs/2603.24639 Come say hi if you are around!
2
11
283
Two months ago, we released ViDoRe V3 benchmark. Now, the full paper is out! 📄arxiv.org/abs/2601.08620 Here is a recap of the benchmark along with a comprehensive breakdown of our findings on multimodal RAG. 🧵 (1/N)
1
9
2
16
721
Victor Xing retweeted
📊 ViDoRe V3 on MTEB: Here we go! The MTEB Leaderboard now includes our latest ViDoRe V3 benchmark, built to rigorously evaluate retrieving in applied and industrial settings. 🤗 Blogpost: huggingface.co/blog/QuentinJ… 🤗 MTEB Leaderboard: huggingface.co/spaces/mteb/l…
1
2
3
406
I'm thrilled to finally announce the release of ViDoRe V3! 🚀 This is our new multimodal retrieval benchmark, built from the ground up for enterprise relevant RAG. We've moving beyond synthetic data to tackle the applied challenges we face every day. (1/N)
📢 ViDoRe V3, our new multimodal retrieval benchmark for enterprise use cases, is finally here! It focuses on real-world applied RAG scenarios using high-quality human-verified data. huggingface.co/blog/QuentinJ… 🧵(1/N)
3
3
4
533
📢 ViDoRe V3, our new multimodal retrieval benchmark for enterprise use cases, is finally here! It focuses on real-world applied RAG scenarios using high-quality human-verified data. huggingface.co/blog/QuentinJ… 🧵(1/N)
5
19
6
75
13,452
1000+ words per second! ⚡ We just unleashed Gemini Diffusion at #GoogleIO! 🚀 Awesome being part of the team that took this from a small research project all the way to I/O @GoogleDeepMind 🪐
We’ve developed Gemini Diffusion: our state-of-the-art text diffusion model. Instead of predicting text directly, it learns to generate outputs by refining noise, step-by-step. This helps it excel at coding and math, where it can iterate over solutions quickly. #GoogleIO
5
30
2
132
9,307