Gate precision: add nationality adjectives + remaining sentence-starters

Gate precision: stop false-positives on art movements / events / starters

Narration: expand per-city gazetteer with famous landmarks (loosen gate)

Call-1 weights: drop the parroted example, default to no-think (A/B verdict)

Call-1 reasoning budget 1024 + no-think A/B flag; narration name lockdown

Call-1 vibe weights: reasoning pass + anti-zero prompt + degenerate guard

Remove stale DEPLOY.md runbook + its dead README link

Submission: finalize field notes + wire all judge links

Narration: remove copyable title examples + deterministic title guard

Narration: stop the 1B model copying the example title verbatim

Narration: kill label-parroting at the source + reduce gate false positives

Narration: stop category-label parroting + force flowing prose

Narration: stop the gate rejecting the LLM's own title + 'Just' (traces-driven)

Trace: create the dataset repo before pushing (upload_file does not auto-create)

Trace: log push failures + boot self-check for HF_TOKEN write access

Narration: route titles, varied phrasing, flowing LLM prose, looser gate

Docs + repo cleanup for submission

UI: style city dropdown to match, condense sliders to one row, live-map teaser loader

new cities added

fixed the narration+traces

Add submission kit: demo script, social post, badge checklist

Docs/UI: reflect multi-city + Off the Grid; Thousand Token Wood track

Pre-baked offline cities (London, Barcelona, New York) + endpoint-stop fix

Mobile UI: bound the autocomplete dropdown + denser drawer + keyboard handling

Fix mode toggle layout: #dr-mode .wrap missing display:flex

Add on-demand worldwide cities (any city via live OSM)

Round-5 fix: relabel bakery_food_shop "food shop" (no false 'bakery' claim)

Round-4 fixes: drop overclaiming match-tag, dedup stops, hidden excludes famous

Round-3 fixes: wire interp weights, ref-name guard, tighten attraction gloss

Round-2 follow-ups: discovery-cue handling + sparse-category honesty

Round-2 route-fit: top-N category cap + weak-match honesty

Route-fit fixes from adversarial review (span, glosses, attraction gate)

Fix Space card: trim short_description to <60 chars, set python_version

Merge pull request #1 from tristanleduc/v2

Merge branch 'v2' of https://github.com/tristanleduc/wanderlust into v2

fixed ui

Import narrate.llm at startup so ZeroGPU detects the @spaces.GPU function

Add src/ path shim so the Space container finds the package

Pin torch 2.11.0 for ZeroGPU compatibility

update readme

v2 with model changed and using gradio server func

Design system, live freshness, autocomplete, and clay-pin markers

Wanderlust: taste-aware Paris discovery routing