eval: agentic results from Actions run [skip ci] f9534ae
github-actions[bot] commited on
Spaces:
Running
Running
Commit History
eval: judge results from Actions run [skip ci] f20fd18
github-actions[bot] commited on
eval: retrieval results from Actions run [skip ci] db5a3dc
github-actions[bot] commited on
retrieval: remove the cross-encoder reranker (ablation showed no earned keep) (#100) 7c37d31 unverified
eliezer avihail Claude commited on
eval: retrieval results from Actions run [skip ci] fc5df43
github-actions[bot] commited on
eval: judge results from Actions run [skip ci] 7abcb44
github-actions[bot] commited on
perf: hydrate retrieved sections concurrently — cut the dominant answer latency (#96) bf7f9a0 unverified
eliezer avihail Claude commited on
eval: judge results from Actions run [skip ci] befd7a2
github-actions[bot] commited on
eval: retrieval results from Actions run [skip ci] bedfc26
github-actions[bot] commited on
eval: built-in rerank ablation, judge on paid hy3, answer-latency measurement (#94) 5bdbb09 unverified
eliezer avihail Claude commited on
eval: judge results from Actions run [skip ci] 5e8f91e
github-actions[bot] commited on
eval: retrieval results from Actions run [skip ci] d338083
github-actions[bot] commited on
eval: docs inventory + index manifest from Actions [skip ci] e8776be
github-actions[bot] commited on
eval: judge results from Actions run [skip ci] 634c9d3
github-actions[bot] commited on
eval: retrieval results from Actions run [skip ci] 6ad0c31
github-actions[bot] commited on
Eval knobs as workflow inputs: rerank on/off and k (#86) 70380bf unverified
eliezer avihail Claude commited on
eval: retrieval results from Actions run [skip ci] 007d14e
github-actions[bot] commited on
eval: docs inventory + index manifest from Actions [skip ci] 2c0f4f5
github-actions[bot] commited on
eval: retrieval results from Actions run [skip ci] 6f4aa37
github-actions[bot] commited on
Answer-quality eval (LLM-judge) + hy3-led OpenRouter default (#81) 0942768 unverified
eliezer avihail Claude commited on
eval: retrieval results from Actions run [skip ci] c580aeb
github-actions[bot] commited on
eval: retrieval results from Actions run [skip ci] 98a9b99
github-actions[bot] commited on
Widen the HNSW candidate scan: kind-filtered dense queries were dropping true neighbors 2b4fdfd unverified
Claude Claude Opus 4.8 commited on
eval: agentic results from Actions run [skip ci] f486f69
github-actions[bot] commited on
Diagnostic: report expected page's true dense rank + distance in the api pool 0a2a35d unverified
Claude Claude Opus 4.8 commited on
eval: retrieval results from Actions run [skip ci] e9d2d82
github-actions[bot] commited on
Merge pull request #69 from eliezeravihail/claude/code-review-improvements-5qn4ke 1e944e0 unverified
eliezer avihail commited on
Make the agentic eval completable: bounded sample + 120m timeout ee4b6fa unverified
Claude Claude Opus 4.8 commited on
Fix per-pool relevance gap: a close tutorial no longer empties the api pool f139f32 unverified
Claude Claude Opus 4.8 commited on
Add one-off retrieval diagnostic: locate expected page in per-kind pools 265382a unverified
Claude Claude Opus 4.8 commited on
eval: retrieval results from Actions run [skip ci] c167ea6
github-actions[bot] commited on
Add the agentic benchmark: coverage via citations, agentic vs single-shot c731ea6 unverified
Claude Claude Fable 5 commited on
Add the v1 eval dataset: 100 valid + 100 invalid + 20 agentic 4b1e499 unverified
Claude Claude Fable 5 commited on
eval: docs inventory + index manifest from Actions [skip ci] 36eb35d
github-actions[bot] commited on
eval: index manifest from Actions run [skip ci] 8297c5c
github-actions[bot] commited on
eval: retrieval v0 results from Actions run [skip ci] 4c5fc95
github-actions[bot] commited on
eval: retrieval v0 results from Actions run [skip ci] 5bb4c35
github-actions[bot] commited on
eval: retrieval v0 results from Actions run [skip ci] 8560150
github-actions[bot] commited on
eval: retrieval v0 results from Actions run [skip ci] 7970cf6
github-actions[bot] commited on
eval: retrieval v0 results from Actions run [skip ci] 36ddc21
github-actions[bot] commited on
Add a retrieval benchmark: recall@k + MRR over the v0 questions 6645570 unverified
Claude Claude Fable 5 commited on
Codebase quality pass: kill silent-truncation hacks, fix latent bugs, cover them (#46) 95d6832 unverified
eliezer avihail Claude commited on
Merge pull request #29 from eliezeravihail/claude/content-extraction-planning-mr59qh dcca7c7 unverified
eliezer avihail commited on
Document M2 grounded before/after: 100% grounded_api_rate, fabrications eliminated 857db73 unverified
Claude Claude Opus 4.8 commited on
eval: v0 results from Actions run [skip ci] fea4d6c
github-actions[bot] commited on
eval: v0 results from Actions run [skip ci] 10c9a11
github-actions[bot] commited on
M2 wiring: grounded answers with validated citations + grounded eval mode 13aab95 unverified
Claude Claude Fable 5 commited on
Close M1: full 15/15 baseline, two fabrication findings, checker fix 421e5bd unverified
Claude Claude Fable 5 commited on
eval: v0 results from Actions run [skip ci] ba2fddd
github-actions[bot] commited on
Document v0 baseline hallucinations: 3 findings from 8 ungrounded answers 6b301a4 unverified
Claude Claude Fable 5 commited on