Nicolas Chapados's Avatar

Nicolas Chapados

@nicolaschapados.bsky.social

Passionate about AI and its impact on society • VP, Research, ServiceNow • Associate industrial member, Mila • Adjunct professor, Polytechnique Montréal • Co-Founder, Imagia & Element AI

180 Followers  |  153 Following  |  5 Posts  |  Joined: 23.11.2024  |  1.3977

Latest posts by nicolaschapados.bsky.social on Bluesky

A new open LLM pushing the boundary of performance vs efficiency, from ServiceNow! 💪🏼✨ 🚀

14.04.2025 07:27 — 👍 1    🔁 0    💬 0    📌 0

Tu connais déjà mon CV 👨‍🔧🙋🏼‍♂️

06.03.2025 00:23 — 👍 2    🔁 0    💬 1    📌 0

Your time series talk was splendid and inspiring, @andrewgwils.bsky.social !

16.12.2024 07:18 — 👍 4    🔁 0    💬 1    📌 0
Post image

We’re really excited to release this large collaborative work for unifying web agent benchmarks under the same roof.

In this TMLR paper, we dive in-depth into #BrowserGym and #AgentLab. We also present some unexpected performances from Claude 3.5-Sonnet

12.12.2024 17:55 — 👍 20    🔁 11    💬 1    📌 2
Post image

🎉 Excited to introduce BigDocs!
An open, transparent multimodal dataset designed for:
📄 Documents
🌐 Web content
🖥️ GUI understanding
👨‍💻 Code generation from images
We’re also launching BigDocs-Bench:
➡️ Document, Web, GUI Visual reasoning
➡️ Converting images into JSON, Markdown, LaTeX, SVG, and more!

10.12.2024 18:34 — 👍 16    🔁 8    💬 1    📌 2
AgentLab diagram.

The image describes AgentLab, a framework for efficient parallel experiments with agents. It highlights:

Core Agent Features:

Dynamic Prompting and a Unified LLM API for interacting with large language models.
BrowserGym Platform:

A tool for testing agents on benchmarks like WebArena, WorkArena, MiniWoB, and others.
Key Features:

Reproducibility, a Unified Leaderboard, an analysis tool called Xray, and a Dataset for sharing agent traces.
Blue elements represent AgentLab components.

AgentLab diagram. The image describes AgentLab, a framework for efficient parallel experiments with agents. It highlights: Core Agent Features: Dynamic Prompting and a Unified LLM API for interacting with large language models. BrowserGym Platform: A tool for testing agents on benchmarks like WebArena, WorkArena, MiniWoB, and others. Key Features: Reproducibility, a Unified Leaderboard, an analysis tool called Xray, and a Dataset for sharing agent traces. Blue elements represent AgentLab components.

🧵-1
We are thrilled to release #AgentLab, a new open-source package for developing and evaluating web agents. This builds on the new #BrowserGym package which supports 10 different benchmarks, including #WebArena.

03.12.2024 21:02 — 👍 18    🔁 15    💬 2    📌 0
INSPIRE

Job opening in experimental particle physics in Montréal. Seems to be a broad search (ATLAS - astrophysics - Dark Matter). Several members of that group also interact with MILA
inspirehep.net/jobs/2841793

26.11.2024 14:58 — 👍 4    🔁 2    💬 0    📌 0

Semantic Scholar is usually pretty good about the final publication venue of a preprint.

25.11.2024 04:15 — 👍 3    🔁 0    💬 1    📌 0

Looking forward to reconnecting with friends and colleagues in Vancouver!

24.11.2024 03:28 — 👍 0    🔁 0    💬 0    📌 0

@nicolaschapados is following 20 prominent accounts