Learning Interpretable Representations Leads to Semantically Faithful EEG-to-Text Generation
This work addresses the reliability issue in generative brain decoding for neuroscience and AI applications, though it appears incremental as it builds on existing methods with new evaluation protocols.
The paper tackled the hallucination problem in EEG-to-text decoding by reframing it as semantic summarization rather than verbatim reconstruction, and proposed the Generative Language Inspection Model (GLIM) which generated fluent, EEG-grounded sentences without teacher forcing on the ZuCo dataset.
Pretrained generative models have opened new frontiers in brain decoding by enabling the synthesis of realistic texts and images from non-invasive brain recordings. However, the reliability of such outputs remains questionable--whether they truly reflect semantic activation in the brain, or are merely hallucinated by the powerful generative models. In this paper, we focus on EEG-to-text decoding and address its hallucination issue through the lens of posterior collapse. Acknowledging the underlying mismatch in information capacity between EEG and text, we reframe the decoding task as semantic summarization of core meanings rather than previously verbatim reconstruction of stimulus texts. To this end, we propose the Generative Language Inspection Model (GLIM), which emphasizes learning informative and interpretable EEG representations to improve semantic grounding under heterogeneous and small-scale data conditions. Experiments on the public ZuCo dataset demonstrate that GLIM consistently generates fluent, EEG-grounded sentences without teacher forcing. Moreover, it supports more robust evaluation beyond text similarity, through EEG-text retrieval and zero-shot semantic classification across sentiment categories, relation types, and corpus topics. Together, our architecture and evaluation protocols lay the foundation for reliable and scalable benchmarking in generative brain decoding.