We show that tiny transformers can profitably employ a simple form of Chain of Thought, which we call protoreasoning, allowing us to study step-by-step reasoning on ~1M-parameter models and opening up opportunities for much more detailed experimentation and analysis than is feasible for larger models. Current Large Language Models exhibit impressive step-by-step reasoning, but we have yet to understand its generality, i.e., when and how LLMs learn genuinely general algorithms rather than"bags of heuristics."Such questions are hard to settle on compute-intensive frontier models trained on opaque data. To work at model scales far below the threshold for natural-language competence, we define reasoning-friendly tasks on Dyck languages (sentences of correctly nested brackets). We find that protoreasoning traces substantially close the out-of-distribution generalization gap, and ablations confirm that the trace's content, not merely its extra tokens, drives the gain.
Chain of Computation (COC) is proposed, a computational architecture that places a transformer-based LM inside an iterative loop, leveraging its strength as a pattern-matching system and can solve TOH problem instances with up to 20 disks, while requiring substantially less training data.
Ehsan Futuhi, Nathan R. Sturtevant· arXiv.org· 0 citations
It is shown that logical incoherencies follow from an LLM’s computation of its internal representations, in particular from an LLM’s failure to take account of the different roles that different expressions may play in determining content.
Nicholas Asher, Swarnadeep Bhar· Topoi· 0 citations
This work demonstrates a concrete failure mode where frontier models exhibit invisible reasoning by leveraging semantically irrelevant filler tokens to improve performance on synthetic reasoning tasks and indicates that frontier models already perform consequential computation with no interpretable trace in their outpu...
Vatsal Baherwani, Tom Goldstein, Ashwinee Panda· arXiv.org· 4 citations· ⚡2
The nature of test-time exploration in RLVR-trained LLMs is investigated by employing controlled maze-solving experiments and extracting a tree structure from mathematical reasoning traces (BODHI-Trees) based on semantic equivalence to delineate between entropy arising from stylistic variations and genuine inferential...
Soumadeep Saha, Krish Sharma, Akshay Chaturvedi et al.· 1 citation
MUX, a simple method for high-bandwidth and compact reasoning based on distillation of discrete reasoning into continuous multiplexed tokens in a latent space, is proposed and suggests that lossless superposition as local learning targets constitutes a sufficient condition for achieving strong and efficient latent cont...
Ayhan Suleymanzade, Halil Alperen Gozeten, Michael M. Bronstein et al.· 1 citation
It is found that, if every rotary component is periodic, RoPE transformers recognize exactly the languages definable in past temporal logic with modular predicates.
Selim Jerad, Anej Svete, Jiaoda Li et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
