Repository navigation
Stop RankingScorer from ranking posts with no Phoenix heads - #29
Closed
Pitchfork-and-Torch wants to merge 19 commits into
Closed
Pitchfork-and-Torch wants to merge 19 commits into
Pitchfork-and-Torch wants to merge 19 commits into
Conversation
in_network_ids is passed to the VF client without deduplication, while oon_ids is deduped four lines below. retweeted_tweet_id is pushed for every candidate that has one, so the same ID repeats once per retweet of a given post — most often when that post is going viral. Neither VfClient implementation dedupes its input: StratoVfClient builds one call per element, and XaiVfClient chunks by XAI_VF_MAX_BATCH_SIZE, so duplicates consume batch slots and can force an extra round trip. Not a correctness issue — results collapse into a HashMap keyed by tweet ID — but redundant work on the For You serving path. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Deduplicate in_network_ids before VF lookup
Missing Phoenix scores used to become 0 and then NEGATIVE_SCORES_OFFSET, so lookup misses and a missing scoring_sequence still made TopK. Leave score unset so TopK drops them. Some(0.0) still ranks. Co-authored-by: Jon Bailey <Pitchfork-and-Torch@users.noreply.github.com>
Owner
Author
|
Wrong base. Ship vehicle is upstream xai-org#142. Do not merge. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Bug
Phoenix lookup miss, missing
scoring_sequence, or overflow pastPHOENIX_CLIENT_MAX_CANDIDATESleaves every Phoenix headNone.RankingScorer::applytreated that as 0, thenoffset_scorewroteSome(NEGATIVE_SCORES_OFFSET).TopKScoreSelectoronly dropsscore == None. Those posts still ranked.Some(0.0)is a real Phoenix prediction and still ranks.This is not #3 (ranker xDS). This is not #31 (author-size IPS). This is not a taste weight tweak.
Five-line proof
PhoenixScorer/predictions.candidate_scoresleave all heads None on missPhoenixCandidatePipelinescorers[PhoenixScorer, RankingScorer, VMRanker]thenTopKScoreSelector(score.unwrap_or(NEG_INFINITY))apply(None) = 0plus offset wrote a finite rank for an unscored postupdate_allalready skipsErr; prediction failure already returnedErrChange
RankingScorerandPhoenixScoresRankingScorer: no heads ->score/weighted_scorestayNonePhoenixScorer: missingscoring_sequencereturnsErrsoupdate_alldoes not stamp empty scores orlast_scored_at_msTests
Some(0.0)favorite: still rankscargo testcannot run. Public dump has no home-mixer crate manifest.Fork PR: none