{"items":[{"id":"c30bb413-02ab-467a-aab7-c0dbbbf367d0","type":"push","org":"Braumeister-Stefan","repo":"lina_database_decoder","title":"Filter out short sequences to improve decipherment reliability","summary":"Added a preprocessing step in the decoder pipeline to discard any database rows containing fewer than two sign groups. By filtering out these inherently less informative or noisy sequences, the frequency analysis and random-init decipherment models can focus on higher-quality, richer data, reducing irrelevant processing. ![Data cleaning time](https://enxegfybrygakxrhnabg.supabase.co/storage/v1/object/public/meme-images/e013f264-3e2a-48ea-8518-6a6893df9fba.jpg)","url":"https://nomit.dev/Braumeister-Stefan/lina_database_decoder/status/4c815748d8a1ae5a23dd2d19820377565bc9760cc9751bf1a4cd5ad8f985e0c0","author":"Braumeister-Stefan","contributors":["Braumeister-Stefan"],"updated_at":"2026-04-12T00:47:59+00:00"},{"id":"4bf6f688-d2d1-41f1-8ee4-0f89568cf117","type":"push","org":"Braumeister-Stefan","repo":"lina_database_decoder","title":"Implemented best-of-N hill-climbing for the 'random-init' decoding strategy.","summary":"Added a new iterative search mechanism to the random-initialization decoding strategy, which now runs $N$ consecutive random cypher generations and retains only the one producing the highest internal semantic consistency score (measured via WordNet). I also introduced a centralized summary output (`outputs/strategy_summary.csv`) to track and rank the performance of all active strategies. This upgrade dramatically improves the quality of random-init outputs and streamlines benchmarking across different approaches.","url":"https://nomit.dev/Braumeister-Stefan/lina_database_decoder/status/8c2555c93c0be8e69402c784379d5a35c9d5f36d82e5a4b6f3c0992f7796da97","author":"Copilot","contributors":["Copilot"],"updated_at":"2026-04-12T00:23:51+00:00"},{"id":"ab74e1e7-5262-47a1-814e-1c5473d6a3a1","type":"push","org":"Braumeister-Stefan","repo":"lina_database_decoder","title":"Boost performance by caching per-token WordNet synsets","summary":"Optimized the semantic consistency scoring process by caching synset lookups on a per-token basis. Previously, the `meaning_scorer` re-queried WordNet synsets for every pairwise comparison, leading to redundant overhead. By shifting to a `synset_map`, we significantly reduce computation time during translation analysis. ![Meaning scores are slow?](https://enxegfybrygakxrhnabg.supabase.co/storage/v1/object/public/meme-images/braumeister-stefan/lina_database_decoder/balloon/71085c83-60f4-456f-ad13-6adc67fa8668.png)","url":"https://nomit.dev/Braumeister-Stefan/lina_database_decoder/status/297d5880626294b8e310c08eb86fcde2d2875bd18b39a822efcbe91538bd24cf","author":"Copilot","contributors":["Copilot"],"updated_at":"2026-04-12T00:14:01+00:00"},{"id":"0248958e-c779-4591-9a1c-be1879df1793","type":"push","org":"Braumeister-Stefan","repo":"lina_database_decoder","title":"Refactored meaning_scorer to use WordNet-based semantic consistency","summary":"We have replaced the dependency on large semantic embedding models for the meaning scorer with a more lightweight, deterministic approach using WordNet's Wu-Palmer similarity metric. This refactor improves maintainability by removing the need for heavy external ML libraries like `sentence-transformers`, while maintaining a robust way to evaluate semantic consistency between translated tokens. ![Codebase refactor](https://enxegfybrygakxrhnabg.supabase.co/storage/v1/object/public/meme-images/5eb2bd94-566d-4d09-a7ad-26d21f262849.png)","url":"https://nomit.dev/Braumeister-Stefan/lina_database_decoder/status/90f36fda5cbc1822c5c115e3dacb3d1aa0ff65466dc1cbf696848c815518cf1e","author":"Copilot","contributors":["Copilot"],"updated_at":"2026-04-12T00:12:34+00:00"},{"id":"9720316a-6b4c-46d5-bab7-3711c7b4da8e","type":"push","org":"Braumeister-Stefan","repo":"lina_database_decoder","title":"Refine suffix stripping logic in meaning_scorer","summary":"Improved the suffix-stripping mechanism in the meaning_scorer's linguistic fallback logic. By reordering the suffix list to prioritize longer matches and simplifying the stripping process, domain vocabulary matching is now more accurate and robust. This improves the quality of meaning assessments when the primary transformer model is unavailable.","url":"https://nomit.dev/Braumeister-Stefan/lina_database_decoder/status/4d9d1b21748450a8b4f48dd4fdccd5b1e0f76c75d063ea7d9263ed3295a8033f","author":"Copilot","contributors":["Copilot"],"updated_at":"2026-04-12T00:01:26+00:00"}],"pagination":{"offset":0,"limit":5,"has_more":true}}