{"paper":{"title":"Escaping the BLEU Trap: A Signal-Grounded Framework with Decoupled Semantic Guidance for EEG-to-Text Decoding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"SemKey grounds EEG-to-text output in neural signals by treating semantic prompts as queries and EEG embeddings as keys and values.","cross_cats":["cs.AI","cs.HC","eess.AS","q-bio.NC"],"primary_cat":"cs.CL","authors_text":"Haonan Wang, Honglong Yang, Xiaomeng Li, Yuchen Wang, Yu Guo","submitted_at":"2026-02-09T02:47:07Z","abstract_excerpt":"Decoding natural language from non-invasive EEG signals is a promising yet challenging task. However, current state-of-the-art models remain constrained by three fundamental issues: Semantic Bias, where outputs collapse into generic linguistic templates; Signal Neglect, where models rely heavily on LLM priors to hallucinate fluent text even in the absence of meaningful signals; and the \"BLEU Trap\", where high-frequency stopwords inflate n-gram metrics, masking a lack of true semantic fidelity. To resolve these challenges, we move beyond conventional end-to-end pipelines and propose SemKey, a n"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"our approach effectively eliminates hallucinations on noise inputs and achieves SOTA performance on these robust protocols","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That injecting semantic prompts as queries and EEG embeddings as key-value pairs will strictly force the LLM to ground generation in neural signals rather than linguistic priors","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SemKey decouples semantic objectives to ground EEG-to-text generation in neural signals, eliminating hallucinations on noise and improving results on retrieval accuracy and Fréchet distance metrics.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"SemKey grounds EEG-to-text output in neural signals by treating semantic prompts as queries and EEG embeddings as keys and values.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"bd259781417e141fd99254fc111a10802b95a808c568b1b582202ee9e9260f4d"},"source":{"id":"2603.03312","kind":"arxiv","version":3},"verdict":{"id":"c2f5201b-000f-4337-91da-23bca34ecc5a","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T06:26:08.101828Z","strongest_claim":"our approach effectively eliminates hallucinations on noise inputs and achieves SOTA performance on these robust protocols","one_line_summary":"SemKey decouples semantic objectives to ground EEG-to-text generation in neural signals, eliminating hallucinations on noise and improving results on retrieval accuracy and Fréchet distance metrics.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That injecting semantic prompts as queries and EEG embeddings as key-value pairs will strictly force the LLM to ground generation in neural signals rather than linguistic priors","pith_extraction_headline":"SemKey grounds EEG-to-text output in neural signals by treating semantic prompts as queries and EEG embeddings as keys and values."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2603.03312/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"2447d0a90e864262fe38d76b61ef463c9120c07435075fbebae6bae0a60f9447"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}