Model releases, architecture work and prompting research on large language models — from frontier-lab announcements to the arXiv papers behind them.
arXiv:2609.23431v1 Announce Type: new Abstract: Human-object interaction (HOI) detection requires grounding an interacting human-object pair and recognizing the verb that links them, often under seve...
arXiv:2603.18532v3 Announce Type: replace-cross Abstract: The strong performance of large vision-language models (VLMs) trained with reinforcement learning (RL) has motivated similar approaches for f...
Apollo Restore is a 24‑billion‑parameter large language model fine‑tuned from Mistral Small to fill in gaps in fragmentary Ancient Greek texts using a fill‑in‑the‑middle objective. It achieves state‑of‑the‑art performance on short lacunae, placing the correct restoration among its top twenty candidates for 80.6% of documentary‑papyrus, 54.6% of literary‑papyrus, and 61.0% of stone‑inscription gaps, outperforming previous models by significant margins. In blind expert evaluations, 20 specialists preferred Apollo Restore over the strongest baseline and judged its performance at least as good as human restorations in 77% of cases, while also improving the published reading of the Vesuvius‑carbonised papyrus P.Herc. 1667.
arXiv:2609.22607v1 Announce Type: new Abstract: We argue here that the current dominant practice in LLM human simulation: prompting instruction-tuned assistant language models to role-play personas,...
arXiv:2609.23153v1 Announce Type: new Abstract: Video diffusion transformers are expensive because attention dominates long spatiotemporal token sequences. We identify the \emph{high-sparsity trap}:...
FlashBoB introduces an I/O‑efficient algorithm for exact backward‑over‑backward (BoB) in softmax attention, enabling precise second‑order differentiation without large intermediate tensors. By exploiting a hierarchical affine structure, the method confines computation to on‑chip tiles and limits off‑chip memory traffic, achieving θ(N² d²/M) HBM usage. Experiments show FlashBoB scales to sequence lengths of 262K on a single A100 GPU, outperforming prior exact baselines and FlashBack by up to 6.3×.
arXiv:2601.15322v3 Announce Type: replace-cross Abstract: Tool-using agents can repeat a final decision while changing their recorded execution. We introduce the Determinism-Faithfulness Assurance Ha...
arXiv:2609.23038v1 Announce Type: cross Abstract: Spatial reasoning is essential for vision-language models (VLMs) to understand and act in the physical world. Reasoning in dynamic environments requi...
arXiv:2609.23465v1 Announce Type: new Abstract: Long-horizon conversational memory is especially challenging in multi-actor settings, where relevant evidence is distributed across participants and co...
arXiv:2609.23999v1 Announce Type: new Abstract: How large language models (LLMs) integrate patient risk with clinical cost tradeoffs remains poorly understood. We investigated how four open-weight LL...
arXiv:2511.04473v3 Announce Type: replace Abstract: Retrieval of information from graph-structured knowledge bases represents a promising direction for improving the factuality of LLMs. While various...
arXiv:2609.22917v1 Announce Type: new Abstract: Non-technical stakeholders frequently cannot write the SQL needed to extract insights from operational databases. We built and evaluated a Text-to-SQL...
Predicting early outcomes based on trajectory can decrease the expenses associated with agent evaluation by terminating a run once the outcome becomes sufficiently predictable, assuming that the predi...
Exhaustively evaluating every candidate LLM configuration on every benchmark item to identify a high-performing one is costly. We formulate configuration selection as a cost-aware Bayesian bandit prob...
We introduce Qwen3.8-Omni-Flash, a natively multimodal agentic model for real-world multimodal productivity. Compared with previous omni models, which primarily emphasized perception and interaction,...
Soccer tactics are interactive: an attacking action changes the opponent's defensive problem, and the observed response depends on the multi-agent match state. We introduce SambaGraph, an action--reac...
TypeSafe AI’s new model, Jev, is a ‘System One’ or decision model that takes text or semi‑structured data as input and outputs floating‑point probabilities for yes/no, choice, or score questions, rather than text. It charges only for input tokens ($0.042/million) and offers free output, making it cheaper and faster than typical LLMs. The API lets users build a state object (string, array, or key‑value pairs) and query it with multiple questions, receiving confidence scores and probability distributions for each answer.
We focus on human-robot collaborative transport, a challenging task of broad relevance spanning logistics, manufacturing, and the home, in which a user and a robot work together to relocate a large or...
Finding circuits in language models usually means running many careful interventions. We try something simpler: treat attention as a routing map from one forward pass, keep a small set of routes that...