AI Safety and Alignment Group at the ELLIS Institute Tübingen and MPI-IS

aisa-group's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Articles

imadreamerboy 

updated 2 collections 27 days ago

Agent Trajectories

Collection

Agent trajectories from the AI Safety and Alignment Group’s benchmarks, released for analysing agent behaviour, capabilities, and safety. • 5 items • Updated 27 days ago

AI Safety Evaluation Datasets

Collection

Datasets and viewer resources for studying evaluation awareness, instrumental behaviour, sabotage, and monitoring in AI agents. • 4 items • Updated 1 day ago

lenalibon 

published a dataset 27 days ago

aisa-group/ResearchArenaMax-Trajectories

Updated 27 days ago • 35 • 3

lenalibon 

updated a dataset 27 days ago

aisa-group/ResearchArenaMax-Trajectories

Updated 27 days ago • 35 • 3

hrdkbhatnagar 

in aisa-group/PostTrainBench-Trajectories about 1 month ago

Remove stale viewer_data files (orphaned by renames + excluded experiments)

#3 opened about 1 month ago by

hrdkbhatnagar

hrdkbhatnagar 

updated a dataset about 1 month ago

aisa-group/PostTrainBench-Trajectories

Updated Aug 6 • 39.4k • 8

hrdkbhatnagar 

in aisa-group/PostTrainBench-Trajectories about 1 month ago

Remove kismet-0715_10h_older_runs (excluded experiment)

#2 opened about 1 month ago by

hrdkbhatnagar

lenalibon 

authored a paper about 1 month ago

ResearchArena: Evaluating Sabotage and Monitoring in Automated AI R&D

Paper • 2607.19321 • Published Jul 21

lenalibon 

updated a dataset about 1 month ago

aisa-group/ResearchArena-Trajectories

Updated Aug 2 • 3.16k • 1

imadreamerboy 

updated 2 collections about 1 month ago

AI Safety Evaluation Datasets

Collection

Datasets and viewer resources for studying evaluation awareness, instrumental behaviour, sabotage, and monitoring in AI agents. • 4 items • Updated 1 day ago

Agent Trajectories

Collection

Agent trajectories from the AI Safety and Alignment Group’s benchmarks, released for analysing agent behaviour, capabilities, and safety. • 5 items • Updated 27 days ago

imadreamerboy 

updated a Space about 1 month ago

Running

AI Safety and Alignment Group

🛡

imadreamerboy 

published a Space about 1 month ago

Running

AI Safety and Alignment Group

🛡

rank-and-file 

published a dataset about 2 months ago

aisa-group/ResearchArena-Trajectories

Updated Aug 2 • 3.16k • 1

rank-and-file 

updated a dataset about 2 months ago

aisa-group/ResearchArena-Trajectories

Updated Aug 2 • 3.16k • 1