model is flan-t5-small

model is flan-t5-small

wip

wip

wip

wip

Reduce log noise from health checks and keep-alive pings

Transform FRONTEND_URL to hf.space/health endpoint and use GET requests

Add info-level log for each keep-alive ping

Fix keep-alive to ping frontend root URL with HEAD request

Add frontend keep-alive pinging

compute consistency

Add consistency score to quantify source/summary alignment

Remove debug API_TOKEN log, add bearer token auth via X-API-Token header

Log API_TOKEN value at startup for debugging

Switch auth to X-API-Token header; HF Spaces strips Authorization

Temporarily disable auth to confirm service works end-to-end

Add auth diagnostic logging to debug 401 mismatch

Simplify bearer token auth: check Authorization header directly in endpoint

Fix bearer token auth: use closure so FastAPI resolves Security() correctly

Add optional bearer token auth to /score endpoint

Include ambiguity_boundaries in normalization block of /score response

Add ambiguity_score to API response with dedicated quantile normalizer

Load scoring service as background task so /is-ready returns false during init

Add GET /is-ready endpoint; load scoring service async in lifespan

Log timestamp when /wake is called

Add GET /wake endpoint for cold-start warm-up

Log quantile boundaries at startup

Remove quantization — incompatible with LoRA weight perturbation

Quantize only non-LoRA layers to avoid perturbation failure

Fix quantization order: apply after PEFT adapter load

Switch back to bart-large-xsum with INT8 quantization for CPU speedup

timer

Wire lora_laplace backend into API server and Dockerfile

Clarify calibration artifacts belong to MC Dropout backend

Align api_server default quantile path with Dockerfile env var

Switch default scoring model to facebook/bart-large-xsum

pass norm boundaries

pass norm boundaries

unc bands

more logs

mc dropout

dummy inf

dummy inf

dummy inf

src