Grgur Kovač
@kovacgrgur.bsky.social
PhD student at INRIA in the Flowers team. https://grgkovac.github.io
Twitter: @KovacGrgur
Twitter: @KovacGrgur
Reposted by Grgur Kovač
What's wrong with evaluating #LLMs after a single interaction? Come find out @iclr-conf.bsky.social and learn how cultural attraction theory can help us do better. Poster #288, 10 am.
April 23, 2025 at 10:11 PM
What's wrong with evaluating #LLMs after a single interaction? Come find out @iclr-conf.bsky.social and learn how cultural attraction theory can help us do better. Poster #288, 10 am.
Reposted by Grgur Kovač
🚀 Introducing 🧭MAGELLAN—our new metacognitive framework for LLM agents! It predicts its own learning progress (LP) in vast natural language goal spaces, enabling efficient exploration of complex domains.🌍✨Learn more: 🔗 arxiv.org/abs/2502.07709 #OpenEndedLearning #LLM #RL
MAGELLAN: Metacognitive predictions of learning progress guide...
Open-ended learning agents must efficiently prioritize goals in vast possibility spaces, focusing on those that maximize learning progress (LP). When such autotelic exploration is achieved by LLM...
arxiv.org
March 24, 2025 at 3:09 PM
🚀 Introducing 🧭MAGELLAN—our new metacognitive framework for LLM agents! It predicts its own learning progress (LP) in vast natural language goal spaces, enabling efficient exploration of complex domains.🌍✨Learn more: 🔗 arxiv.org/abs/2502.07709 #OpenEndedLearning #LLM #RL
LLama 3.3 is great, but Nemotron is still the leader in our StickToYourRole Leaderboard !
Nemotron 🥇
Llama 3.3 🥈
huggingface.co/spaces/flowe...
Nemotron 🥇
Llama 3.3 🥈
huggingface.co/spaces/flowe...
December 10, 2024 at 2:15 PM
LLama 3.3 is great, but Nemotron is still the leader in our StickToYourRole Leaderboard !
Nemotron 🥇
Llama 3.3 🥈
huggingface.co/spaces/flowe...
Nemotron 🥇
Llama 3.3 🥈
huggingface.co/spaces/flowe...
Reposted by Grgur Kovač
I'm excited to announce that this work has been accepted at
@blog.neurips.cc.web.brid.gy 🧠🤖 We hope to spark conversations on goal selection in biological and artificial agents.
Check it out at openreview.net/forum?id=Gbq...
With Cédric Colas, Pierre-Yves Oudeyer, & Anne Collins
@blog.neurips.cc.web.brid.gy 🧠🤖 We hope to spark conversations on goal selection in biological and artificial agents.
Check it out at openreview.net/forum?id=Gbq...
With Cédric Colas, Pierre-Yves Oudeyer, & Anne Collins
November 18, 2024 at 8:20 PM
I'm excited to announce that this work has been accepted at
@blog.neurips.cc.web.brid.gy 🧠🤖 We hope to spark conversations on goal selection in biological and artificial agents.
Check it out at openreview.net/forum?id=Gbq...
With Cédric Colas, Pierre-Yves Oudeyer, & Anne Collins
@blog.neurips.cc.web.brid.gy 🧠🤖 We hope to spark conversations on goal selection in biological and artificial agents.
Check it out at openreview.net/forum?id=Gbq...
With Cédric Colas, Pierre-Yves Oudeyer, & Anne Collins
Reposted by Grgur Kovač
🚨New preprint🚨
When testing LLMs with questions, how can we know they did not see the answer in their training? In this new paper we propose a simple out of the box and fast method to spot contamination on short texts with @stepalminteri.bsky.social and Pierre-Yves Oudeyer !
When testing LLMs with questions, how can we know they did not see the answer in their training? In this new paper we propose a simple out of the box and fast method to spot contamination on short texts with @stepalminteri.bsky.social and Pierre-Yves Oudeyer !
November 15, 2024 at 1:48 PM
🚨New preprint🚨
When testing LLMs with questions, how can we know they did not see the answer in their training? In this new paper we propose a simple out of the box and fast method to spot contamination on short texts with @stepalminteri.bsky.social and Pierre-Yves Oudeyer !
When testing LLMs with questions, how can we know they did not see the answer in their training? In this new paper we propose a simple out of the box and fast method to spot contamination on short texts with @stepalminteri.bsky.social and Pierre-Yves Oudeyer !