220
308
Understanding Generalization in Role-Playing Models via Information Theory
arXiv:2512.17270v1 Announce Type: new
Abstract: Role-playing models (RPMs) are widely used in real-world applications but underperform when deployed in the wild. This degradation can be attributed to distribution shifts, including user, character, and dialogue compositional shifts. Existing methods…
Abstract: Role-playing models (RPMs) are widely used in real-world applications but underperform when deployed in the wild. This degradation can be attributed to distribution shifts, including user, character, and dialogue compositional shifts. Existing methods…
Maurozuhause
320
Structure-Aware Antibody Design with Affinity-Optimized Inverse Folding
arXiv:2512.17815v1 Announce Type: new
Abstract: Motivation: The clinical efficacy of antibody therapeutics critically depends on high-affinity target engagement, yet laboratory affinity-maturation campaigns are slow and costly. In computational settings, most protein language models (PLMs) are not …
Abstract: Motivation: The clinical efficacy of antibody therapeutics critically depends on high-affinity target engagement, yet laboratory affinity-maturation campaigns are slow and costly. In computational settings, most protein language models (PLMs) are not …
Shelly
208
Russia plans to revive abandoned Soviet-era particle accelerator
Accelerator could be used to generate an intense beam of neutrinos
The post Russia plans to revive abandoned Soviet-era particle accelerator appeared first on Physics World.
The post Russia plans to revive abandoned Soviet-era particle accelerator appeared first on Physics World.
Fasching77
120
A Benchmark and Agentic Framework for Omni-Modal Reasoning and Tool Use in Long Videos
arXiv:2512.16978v1 Announce Type: new
Abstract: Long-form multimodal video understanding requires integrating vision, speech, and ambient audio with coherent long-range reasoning. Existing benchmarks emphasize either temporal length or multimodal richness, but rarely both and while some incorporate…
Abstract: Long-form multimodal video understanding requires integrating vision, speech, and ambient audio with coherent long-range reasoning. Existing benchmarks emphasize either temporal length or multimodal richness, but rarely both and while some incorporate…
Barcresatzrek
119
meval: A Statistical Toolbox for Fine-Grained Model Performance Analysis
arXiv:2512.17409v1 Announce Type: new
Abstract: Analyzing machine learning model performance stratified by patient and recording properties is becoming the accepted norm and often yields crucial insights about important model failure modes. Performing such analyses in a statistically rigorous manne…
Abstract: Analyzing machine learning model performance stratified by patient and recording properties is becoming the accepted norm and often yields crucial insights about important model failure modes. Performing such analyses in a statistically rigorous manne…
Jasonbraun26
234
Feedback Arc Sets and Feedback Arc Set Decompositions in Weighted and Unweighted Oriented Graphs
arXiv:2501.06935v3 Announce Type: replace-cross
Abstract: Let $D=(V(D),A(D))$ be a digraph with at least one directed cycle. A set $F$ of arcs is a feedback arc set (FAS) if $D-F$ has no directed cycle. The FAS decomposition number ${\rm fasd}(D)$ of $D$ is the maximum number of pairwise disjoint F…
Abstract: Let $D=(V(D),A(D))$ be a digraph with at least one directed cycle. A set $F$ of arcs is a feedback arc set (FAS) if $D-F$ has no directed cycle. The FAS decomposition number ${\rm fasd}(D)$ of $D$ is the maximum number of pairwise disjoint F…
Jpatrick2682
317
Universal consistency of the $k$-NN rule in metric spaces and Nagata dimension. III
arXiv:2512.17058v1 Announce Type: new
Abstract: We prove the last remaining implication allowing to claim the equivalence of the following conditions for a complete separable metric space $X$:
(1) The $k$-nearest neighbour classifier is (weakly) universally consistent in $X$, (2) The strong Lebes…
Abstract: We prove the last remaining implication allowing to claim the equivalence of the following conditions for a complete separable metric space $X$:
(1) The $k$-nearest neighbour classifier is (weakly) universally consistent in $X$, (2) The strong Lebes…
Chad0999
220
Probing Scientific General Intelligence of LLMs with Scientist-Aligned Workflows
arXiv:2512.16969v1 Announce Type: new
Abstract: Despite advances in scientific AI, a coherent framework for Scientific General Intelligence (SGI)-the ability to autonomously conceive, investigate, and reason across scientific domains-remains lacking. We present an operational SGI definition grounde…
Abstract: Despite advances in scientific AI, a coherent framework for Scientific General Intelligence (SGI)-the ability to autonomously conceive, investigate, and reason across scientific domains-remains lacking. We present an operational SGI definition grounde…
Swelding407
221
The Ten Most Significant Science Stories of 2025
Kranking, Prillaman, & Hill, Sonian Every December, Smithsonian magazine's science team picks which stories were the year's most significant. We look for a broad set of topics that...
Lylerocker33
109
Extended formulations for induced tree and path polytopes of chordal graphs
arXiv:2512.08554v2 Announce Type: replace
Abstract: In this article, we give two extended space formulations, respectively, for the induced tree and path polytopes of chordal graphs with vertex and edge variables.
These formulations are obtained by proving that the induced tree and path extended …
Abstract: In this article, we give two extended space formulations, respectively, for the induced tree and path polytopes of chordal graphs with vertex and edge variables.
These formulations are obtained by proving that the induced tree and path extended …
Lpike
111
Explorable Ideas: Externalizing Ideas as Explorable Environments
arXiv:2512.17017v1 Announce Type: new
Abstract: Working with abstract information often relies on static, symbolic representations that constrain exploration. We introduce Explorable Ideas, a framework that externalizes abstract concepts into explorable environments where physical navigation coordi…
Abstract: Working with abstract information often relies on static, symbolic representations that constrain exploration. We introduce Explorable Ideas, a framework that externalizes abstract concepts into explorable environments where physical navigation coordi…
Dkeleti
109
Utility-Diversity Aware Online Batch Selection for LLM Supervised Fine-tuning
arXiv:2510.16882v2 Announce Type: replace
Abstract: Supervised fine-tuning (SFT) is a commonly used technique to adapt large language models (LLMs) to downstream tasks. In practice, SFT on a full dataset is computationally expensive and sometimes suffers from overfitting or bias amplification. This…
Abstract: Supervised fine-tuning (SFT) is a commonly used technique to adapt large language models (LLMs) to downstream tasks. In practice, SFT on a full dataset is computationally expensive and sometimes suffers from overfitting or bias amplification. This…
Gibby954
220
Refinement-based Christoffel sampling for least squares approximation in non-orthogonal bases
arXiv:2510.08461v2 Announce Type: replace
Abstract: We introduce a refinement-based Christoffel sampling (RCS) algorithm for least squares approximation in the span of a given, generally non-orthogonal set of functions $\Phi_n = \{\phi_1, \dots, \phi_n\}$. A standard sampling strategy for this prob…
Abstract: We introduce a refinement-based Christoffel sampling (RCS) algorithm for least squares approximation in the span of a given, generally non-orthogonal set of functions $\Phi_n = \{\phi_1, \dots, \phi_n\}$. A standard sampling strategy for this prob…
Tim_moore001
234
Strategic Planning and Rationalizing on Trees Make LLMs Better Debaters
arXiv:2505.14886v2 Announce Type: replace
Abstract: Winning competitive debates requires sophisticated reasoning and argument skills. There are unique challenges in the competitive debate: (1) The time constraints force debaters to make strategic choices about which points to pursue rather than cov…
Abstract: Winning competitive debates requires sophisticated reasoning and argument skills. There are unique challenges in the competitive debate: (1) The time constraints force debaters to make strategic choices about which points to pursue rather than cov…
Joe S
109
Alzheimer's Disease Brain Network Mining
arXiv:2512.17276v1 Announce Type: new
Abstract: Machine learning approaches for Alzheimer's disease (AD) diagnosis face a fundamental challenges. Clinical assessments are expensive and invasive, leaving ground truth labels available for only a fraction of neuroimaging datasets. We introduce Multi v…
Abstract: Machine learning approaches for Alzheimer's disease (AD) diagnosis face a fundamental challenges. Clinical assessments are expensive and invasive, leaving ground truth labels available for only a fraction of neuroimaging datasets. We introduce Multi v…
Qwertass
423
Ancient sea anemone sheds light on animal cell type evolution
One of the biggest quests in biology is understanding how every cell in an animal's body carries an identical genome yet still gives rise to a kaleidoscope of different cell types and tissues. A neuron doesn't look nor behave like a muscle cell but has the same DNA.
Gthielens
111
Robust-R1: Degradation-Aware Reasoning for Robust Visual Understanding
arXiv:2512.17532v1 Announce Type: new
Abstract: Multimodal Large Language Models struggle to maintain reliable performance under extreme real-world visual degradations, which impede their practical robustness. Existing robust MLLMs predominantly rely on implicit training/adaptation that focuses sol…
Abstract: Multimodal Large Language Models struggle to maintain reliable performance under extreme real-world visual degradations, which impede their practical robustness. Existing robust MLLMs predominantly rely on implicit training/adaptation that focuses sol…
Gibby954
210
Deterministic implementation in single-item auctions
arXiv:2512.17386v1 Announce Type: new
Abstract: Deterministic auctions are attractive in practice due to their transparency, simplicity, and ease of implementation, motivating a sharp understanding of when they match randomized designs. We study deterministic implementation in single-item auctions …
Abstract: Deterministic auctions are attractive in practice due to their transparency, simplicity, and ease of implementation, motivating a sharp understanding of when they match randomized designs. We study deterministic implementation in single-item auctions …
Unknown
120
More Consistent Accuracy PINN via Alternating Easy-Hard Training
arXiv:2512.17607v1 Announce Type: new
Abstract: Physics-informed neural networks (PINNs) have recently emerged as a prominent paradigm for solving partial differential equations (PDEs), yet their training strategies remain underexplored. While hard prioritization methods inspired by finite element …
Abstract: Physics-informed neural networks (PINNs) have recently emerged as a prominent paradigm for solving partial differential equations (PDEs), yet their training strategies remain underexplored. While hard prioritization methods inspired by finite element …
Kathy