Repository navigation
Reports: named files vote in where_to_look (dev Acc@1 0.249 → 0.298); LocAgent head-to-head; paper draft - #137
Merged
Merged
Conversation
…c@1 0.249 -> 0.298) A traceback's frames (deepest first) and the paths / unique file names a report mentions, resolved against the index by longest unique path suffix, join the long-report fusion as a third RRF vote at weight 2. SWE-bench dev (225): Acc@1/3/5/10 0.249/0.484/0.600/0.680 -> 0.298/0.516/0.609/0.689 (prototype); both halves of dev agree; the engine port reproduces it on dev-fast. Fourteen sets unchanged. Also: LocAgent's exact 274-instance Lite subset rebuilt (head-to-head: 0.474 / 0.708 / 0.752 vs Jina-Code-v2 0.434 / 0.712 / 0.803 and Agentless+GPT-4o @5 0.745), research scripts for a local-LLM stage (llama.cpp), the roadmap to the paper and a living paper draft. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
locagent_subset.py): v1.2.0 0.474/0.708/0.752 vs Jina-Code-v2 0.434/0.712/0.803, Agentless+GPT-4o @5 0.745.13-roadmap-paper.fa.md,paper-draft.md.Fourteen benchmark sets unchanged.
🤖 Generated with Claude Code