Don't Train the Model, Evolve the Harness
🌿
1
Evolving an agent's harness, not its model, on Harvey's LAB
Joel Niklaus
joelniklaus
AI & ML interests
Pretraining, Instruction Tuning, Domain Adaptation, Benchmarks, Legal Datasets
Recent Activity
published a Space 6 days ago
joelniklaus/harness-optimization updated a Space 6 days ago
joelniklaus/harness-optimization updated a Space 9 days ago
HuggingFaceFW/finephraseOrganizations
published a
Space 6 days ago
updated a
Space 6 days ago
updated a
Space 9 days ago
upvoted an article 14 days ago
Article
PhysicsIntern: from an Autonomous Benchmark-runner to a Research Sidekick
dlouapre
• • 6 updated 11
models 16 days ago
verify-ppt/marin_3b
3B • Updated • 42
verify-ppt/smollm3_3b
3B • Updated • 89
verify-ppt/olmo3_3b
3B • Updated • 120
verify-ppt/c4_3b
3B • Updated • 48
verify-ppt/ppt-set_c4_3b_v2
3B • Updated • 48
verify-ppt/ppt_c4_3b_v2
3B • Updated • 54
verify-ppt/ppt_marin_3b_v2
3B • Updated • 53
verify-ppt/ppt_olmo3_3b_v2
3B • Updated • 45
verify-ppt/ppt_smollm3_3b_v2
3B • Updated • 89
verify-ppt/ppt-set_3b_v2
3B • Updated • 13
verify-ppt/ppt_3b_v2
3B • Updated • 9
upvoted an article 17 days ago
Article
Designing the hf CLI as an agent-optimized way to work with the Hub
celinah, Wauplin
• • 58 updated a
bucket 21 days ago
Add LEXam evaluation results
#161 opened 23 days ago
by joelniklaus
Add LEXam evaluation results
#48 opened 23 days ago
by joelniklaus
Add LEXam evaluation results
#34 opened 23 days ago
by joelniklaus