fix: pre-translate non-English chunks to English before LLM inference

revert: restore original demo.launch(), blog served from separate static Space

fix: inject /blog into Gradio's internal FastAPI router at position 0

fix: mount /blog before Gradio so static files aren't swallowed by catch-all route

feat: serve blog at /blog from main Space + badge showcase

feat: full-height libraries, pre-generated reward image, English upload text

fix: make the llama.cpp CPU runtime actually usable on the Space

fix: library panels full width, hidden rows off-screen

feat: animated library side panels with mini robots

feat: amphitheater hero with mouse-tracking robot professor, fixed particles

fix: MCQ clicks broken — restore onclick in showMCQ, remove duplicate addEventListener

fix: MCQ double-count — remove duplicate onclick handler in showMCQ

feat: FLUX image generation moved to End Session modal as session visual summary

fix: increase GPU duration to 120s, prefetch FLUX weights on startup

fix: FLUX.2-klein prompt keyword arg, image for all verdicts

feat: FLUX.1-schnell image generation after correct answers

ui: replace mode pills with large cards (icon, title, desc, check indicator)

fix: English only, remove language toggle

fix: MCQ onclick via closure, add FR/EN language pills wired to hidden textbox

feat: add MCQ mode with 4 options, client-side evaluation and per-choice explanations

fix: truncate source passage to 3 sentences / 400 chars for display

fix: expose chunk via hidden textbox so source passage displays in feedback

fix: replace hidden button clicks with textbox triggers (setGradioTA works, btn.click() doesn't in Gradio SSR)

fix: replace display:none on hidden rows with height:0 so Gradio mounts event handlers

debug: add prints in new_question to trace call and LLM progress

fix: elapsed time counter + 3min hard timeout on question generation

fix: infinite spinner on question gen, 4-bit quant for low VRAM

fix: full quiz flow via BRIDGE_JS, English only, buttons DOM-visible

fix: move bridge JS to demo.load() script in gr.HTML never executed

fix: hero visible, polling replaces MutationObserver for Svelte compat

fix: hero visible, upload row vertical, thin lines gone

fix: hero HTML fragment, invisible textboxes, gradient button

fix: make status/question/feedback textboxes DOM-visible for JS observer

feat: beautiful dark theme UI with smart status bridge

feat: hybrid approach native Gradio file upload styled with custom CSS

fix: add explicit demo.launch() with server_name and port

fix: remove if __name__ block so HF Spaces auto-launches demo

debug: add instrumentation to find crash location

fix: suppress warnings, css back in gr.Blocks for HF Spaces

feat: Off-Brand badge - custom HTML UI with JS bridge to hidden Gradio components

feat: switch to Docker SDK with custom FastAPI frontend at /

fix: back to app.py entry point, custom UI at /paperprof-ui

fix: use @spaces.GPU directly, remove theme/css from gr.Blocks constructor

feat: Off-Brand badge - custom FastAPI+HTML frontend, zero Gradio chrome

fix: même fond sombre pour status/question/feedback que Your Answer

fix: texte question/feedback invisible + prompt langue renforcé

fix: langue par défaut English

fix: retire primary_hue orange qui changeait toutes les teintes

fix: revenir au style sombre Soft, juste boutons colorés + zone upload propre

fix: revenir à Soft theme (file upload correct) + override cards blanches