CoolFace
Apppublic

flavianv/reco-protocol-a-leaderboard

sourceHugging Faceupdated 6mo agoView on Hugging Face
0likes
50 commits on main
ac431666mo ago

Add paired SmartBench Qwen runs

Flavian Vasile
0e507756mo ago

Report SmartBench partial query progress

Flavian Vasile
741e45a6mo ago

Add logging to SmartBench Qwen probe

Flavian Vasile
d709d8b6mo ago

Fix SmartBench HF dependency versions

Flavian Vasile
db07c376mo ago

Add SmartBench Qwen XS probe script

Flavian Vasile
fc4e4b16mo ago

Prewarm SmartBench Qwen assets on startup

Flavian Vasile
a6ba1c26mo ago

Stream SmartBench HF asset downloads

Flavian Vasile
440b2ae6mo ago

Add SmartBench Qwen retrieval toolsets

Flavian Vasile
4b901a06mo ago

Show neutral PCR for semantic SmartBench runs

Flavian Vasile
64d90eb6mo ago

Allow single-round SmartBench search runs

Flavian Vasile
d1b0fa26mo ago

Fix SmartBench retrieval cold-start search

Flavian Vasile
677e92f6mo ago

Add SmartBench perf PCR controls

Flavian Vasile
8bfbd416mo ago

Persist SmartBench perf subscore fields

Flavian Vasile
6d190176mo ago

Backfill missing SmartBench table subscores

Flavian Vasile
769e74e6mo ago

Show SmartBench subscores in eval tables

Flavian Vasile
bc421256mo ago

Always score SmartBench avg_perf with perf backend

Flavian Vasile
c3ee5ca6mo ago

Use retrieval scores for SmartBench avg_perf

Flavian Vasile
3eb96b56mo ago

Tune SmartBench perf score weights

Flavian Vasile
eb542846mo ago

Fix SmartBench perf query scoring path

Flavian Vasile
a06671f6mo ago

Load SmartBench recsys E5 model via PEFT

Flavian Vasile
17309ba6mo ago

Switch SmartBench to shared E5 retrieval artifacts

Flavian Vasile
675784f6mo ago

Label active SmartBench agent prompt mode

Flavian Vasile
72ae7aa6mo ago

Clarify updated SmartBench perf toolset UI

Flavian Vasile
8f2685c6mo ago

Add SmartBench perfmodel search backend

Flavian Vasile
229daa16mo ago

Persist SmartBench run progress and partial logs

Flavian Vasile
7ac54016mo ago

Default SmartBench runs per query to one

Flavian Vasile
9f50e546mo ago

Show seconds in SmartBench duration columns

Flavian Vasile
9d5d97e6mo ago

Add SmartBench evaluation duration metrics

Flavian Vasile
023da6d6mo ago

Fix SmartBench HF musical asset source

Flavian Vasile
42c97886mo ago

Refine SmartBench perf complements and deduping

Flavian Vasile
7fb268e6mo ago

Remove popularity from semantic search output

Flavian Vasile
e8394976mo ago

Simplify perf search reranking

Flavian Vasile
7980d236mo ago

Filter non-agent HF chat model candidates

Flavian Vasile
af1bb026mo ago

Loosen HF chat model discovery filter

Flavian Vasile
28351176mo ago

Trim unsupported HF model presets

Flavian Vasile
7dbeb186mo ago

Refresh model presets from small-model shortlist

Flavian Vasile
d96a0e06mo ago

Backfill failure columns for legacy leaderboard rows

Flavian Vasile
8390bd46mo ago

Format eval ETA as h m s

Flavian Vasile
51c31e06mo ago

Add model preset dropdown

Flavian Vasile
68111cd6mo ago

Count agent signature violations in eval stats

Flavian Vasile
d1aba266mo ago

Surface eval failures in main panel

Flavian Vasile
8c768366mo ago

Show eval progress in main panel

Flavian Vasile
afc16dd6mo ago

Show query progress and ETA during eval

Flavian Vasile
3c570a46mo ago

Show live eval progress in app

Flavian Vasile
f16b3fc6mo ago

Load local env and reset SmartBench state

Flavian Vasile
002e3816mo ago

Disable SmartBench agent fallback

Flavian Vasile
4f162746mo ago

Reorder leaderboard tables and add query counts

Flavian Vasile
3315a976mo ago

Rename metric definitions page to metrics

Flavian Vasile
0812f5c6mo ago

Accept sized HF musical asset sources

Flavian Vasile
e8875d56mo ago

Materialize HF musical pack query files

Flavian Vasile