Recursive reasoning models apply a small shared Transformer block many times to refine a latent state. This gives them large effective depth with few parameters and makes them strong on algorithmic tasks. Such compact solvers are natural candidates for tools that an LLM can call on narrow algorithmic subproblems. Howev...
Yuliana Shakhvalieva, Dmitrii Kharchev, В. Е. Безруков et al.· 0 citations
Masked diffusion models (MDMs) generate sequences by progressively unmasking several tokens per denoising step, but their reverse process is typically factorized over positions, limiting sample quality in the few-step regime where diffusion's speed advantage over autoregressive decoding matters most. A recent line of w...
Arseny Ivanov, A. Kolesov, Alexander Korotin et al.· 0 citations
Language models can now prove theorems, but people still decide which problems to pursue. We ask whether a model's internal representations can help identify promising mathematical connections. We develop LANTERN, a fast, cost-efficient pipeline that uses a classifier over pretrained-model activations to rank candidate...
Pavel Tikhonov, Elena Tutubalina, I. Oseledets et al.· 0 citations
This work constructs a family with quadratic arithmetic in the matrix dimension when the physical block size remains fixed, and derive finite-shape constraints for GPU execution, and treats these results as a feasibility and trainability check for the proposed approach at small scale, leaving further investigation to f...
Geometric Fixed-Point Reasoning (GFPR), in which the iterated state is the prediction itself: a field of categorical beliefs on a product of simplices, whose argmax is the answer at every step.
T. Daulbaev, И. Д. Глазков, Maxim V. Rakhuba et al.· 0 citations
The results show that model size strongly affects generation quality, multimodal training improves the use of individual modalities, and multimodal inference provides a stronger contextual signal than isolated context inputs.
Evgenii Maslov, Alexandra Vabnits, Vladimir Vorona et al.· Smart Cities· 0 citations
We introduce methods to quantify how Large Language Models (LLMs) encode and store contextual information, revealing that tokens often seen as minor (e.g., determiners, punctuation) carry surprisingly high context. Notably, removing these tokens -- especially stopwords, articles, and commas -- consistently degrades per...
Anton Razzhigaev, Matvey Mikhalchuk, Temur Rahmatullaev et al.· North American Chapter of th...· 18 citations· ⚡2
A gap between algorithmic losslessness and its implementation under finite-precision arithmetic is demonstrated and motivated, to motivate evaluating lossless speculative decoding at the level of exact generation trajectories as well as downstream task performance.
Ilya Koziev, Leonid S. Sinev, I. Oseledets· 0 citations
The proposed greedy algorithm with alternating updates guarantees the four-peak distribution required for stable 2-bit clustering of each factor and admits closed-form initialization of cluster centers, removing the multi-restart k-means bottleneck of prior work.
Daria Cherniuk, A. Rudikov, B. Kashin et al.· 0 citations
Modified SINNs are introduced, which integrate coefficient decay scaling and basis embeddings motivated by harmonic analysis to enhance accuracy in high-dimensional problems and enable accurate approximation of unknown spectral coefficients.
Pyligent, a training and inference framework inspired by the Diligent Learner formulation that represents reasoning as validated search over partial solution chains, is introduced and results suggest that explicit failed-branch supervision can teach useful recovery behavior beyond imitation of polished solution chains.
Dmitry N. Beresnev, Vladimir Makharev, Roman Khalikov et al.· 1 citation
EvoMem is introduced, a persistent memory architecture for LLM-based evolutionary program search that captures and reuses candidate mutation knowledge and provides evidence that persistent memory can reduce some redundant exploration and improve the reuse and adaptation of successful strategies in LLM-driven evolutiona...
Viktor Volkov, Valentin Khrulkov, Andrey V. Galichin 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.
