Blog on RTG Neuroexplicit Models https://www.neuroexplicit.org/blog/ Recent content in Blog on RTG Neuroexplicit Models Hugo en-us Mon, 10 Aug 10100 12:00:00 +0200 Temporal Context Reinstatement in Long-Context Language Models https://www.neuroexplicit.org/blog/temporal-context-reinstatement/ Mon, 10 Aug 10100 12:00:00 +0200 https://www.neuroexplicit.org/blog/temporal-context-reinstatement/ <p>Paper: <a href="https://arxiv.org/abs/2607.22575">arXiv</a>, ICML'26 | Software: <a href="https://github.com/mathispink/temporal-context-reinstatement">GitHub</a> | Keywords: <em>interpretability, long context, attention heads, causal interventions</em></p> Retrieving Semantics from the Deep: an RAG Solution for Gesture Synthesis https://www.neuroexplicit.org/blog/2026-07-15-rag-gesture-blog/ Wed, 08 Jul 8080 12:00:00 +0200 https://www.neuroexplicit.org/blog/2026-07-15-rag-gesture-blog/ <p>Paper: <a href="https://arxiv.org/abs/2412.06786">arXiv</a>, CVPR'25 | Software: <a href="https://github.com/m-hamza-mughal/RAG-Gesture.git">GitHub</a> | Keywords: <em>co-speech gestures, diffusion models, retrieval-augmented generation, neuro-explicit</em></p> Discovering Interpretable Algorithms by Decompiling Transformers to RASP https://www.neuroexplicit.org/blog/2026-06-10-decompiling-transformers/ Wed, 10 Jun 10100 12:00:00 +0200 https://www.neuroexplicit.org/blog/2026-06-10-decompiling-transformers/ <p>Paper: <a href="https://arxiv.org/abs/2602.08857">arXiv</a>, ICML'26 | Software: <a href="https://github.com/lacoco-lab/decompiling_transformers">GitHub</a> | Keywords: <em>NLP, interpretability, transformers</em></p> Discover-then-name: Task-agnostic concept bottlenecks via automated concept discovery https://www.neuroexplicit.org/blog/2026-01-16/ Fri, 16 Jan 16160 12:00:00 +0100 https://www.neuroexplicit.org/blog/2026-01-16/ <p>Paper: <a href="https://link.springer.com/chapter/10.1007/978-3-031-72980-5_26">ECCV'24</a> | Software: <a href="https://github.com/neuroexplicit-saar/Discover-then-Name">GitHub</a> | Keywords: <em>CV, concept bottleneck models</em></p> RSA-Control: A Pragmatics-Grounded Lightweight Controllable Text Generation Framework https://www.neuroexplicit.org/blog/2025-12-15/ Mon, 15 Dec 15150 02:00:00 +0100 https://www.neuroexplicit.org/blog/2025-12-15/ <p>Paper: <a href="https://aclanthology.org/2024.emnlp-main.318">EMNLP'24</a> | Software: <a href="https://github.com/Ewanwong/RSA-Control">GitHub</a> | Keywords: <em>NLP, controlled generation, rational speech act</em></p> Introducing the Neuroexplicit Models Blog https://www.neuroexplicit.org/blog/2025-11-14/ Fri, 14 Nov 14140 12:00:00 +0100 https://www.neuroexplicit.org/blog/2025-11-14/ <!-- Keywords: _Neurosymbolic, AI, ML_ -->