Mafia is a social deduction game where informed mafia compete against uninformed townsfolk. Its asymmetry of information and reliance on theory-of-mind reasoning mirror real-world multi-agent scenarios, making it a useful testbed for evaluating the social intelligence of large language models (LLMs). To support a systematic study, we introduce Mini-Mafia: a simplified four-player variant with one mafioso, one detective, and two villagers. We set the mafioso to kill a villager and the detective to investigate the mafioso during the night, reducing the game to a single day phase of discussion and voting. This setup isolates three interactive capabilities through role-specific win conditions: the mafioso must deceive, the villagers must detect deception, and the detective must effectively disclose information. To measure these skills, we have LLMs play against each other, creating the Mini-Mafia Benchmark: a two-stage framework that first estimates win rates within fixed opponent configurations, then aggregates performance across them using standardized scoring. Built entirely from model interactions without external data, the benchmark evolves as new models are introduced, with each one serving both as a new opponent and as a subject of evaluation. Our experiments reveal counterintuitive results, including cases where smaller models outperform larger ones. Beyond benchmarking, Mini-Mafia enables quantitative study of emergent multi-agent dynamics such as name bias and last-speaker advantage. It also contributes to AI safety by generating training data for deception detectors and by tracking models' deception capabilities against human baselines.

Recent work in Topological Deep Learning (TDL) seeks to generalize graph learning's preeminent $message \ passing$ paradigm to more complex relational structures: simplicial complexes, cell complexes, hypergraphs, and combinations thereof. Many approaches to such ${higher\text{-}order \ message \ passing}$ (HOMP) admit formulation in terms of nonlinear diffusion with the Hodge (combinatorial) Laplacian, a graded operator which carries an inductive bias that dimension-$k$ data features correlate with dimension-$k$ topological features encoded in the (singular) cohomology of the underlying domain. For $k=0$ this recovers the graph Laplacian and its well-studied homophily bias. In higher gradings, however, the Hodge Laplacian's bias is more opaque and potentially even degenerate. In this essay, we position sheaf theory as a natural and principled formalism for modifying the Hodge Laplacian's diffusion-mediated interface between local and global descriptors toward more expressive message passing. The sheaf Laplacian's inductive bias correlates dimension-$k$ data features with dimension-$k$ $sheaf$ cohomology, a data-aware generalization of singular cohomology. We will contextualize and novelly extend prior theory on sheaf diffusion in graph learning ($k=0$) in such a light -- and explore how it fails to generalize to $k>0$ -- before developing novel theory and practice for the higher-order setting. Our exposition is accompanied by a self-contained introduction shepherding sheaves from the abstract to the applied.

Agentic AI represents a paradigm shift in enhancing the capabilities of generative AI models. While these systems demonstrate immense potential and power, current evaluation techniques primarily focus on assessing their efficacy in identifying appropriate agents, tools, and parameters. However, a critical gap exists in evaluating the alignment between an Agentic AI system's tasks and its overarching goals. This paper introduces the Creative Adversarial Testing (CAT) framework, a novel approach designed to capture and analyze the complex relationship between Agentic AI tasks and the system's intended objectives. We validate the CAT framework through extensive simulation using synthetic interaction data modeled after Alexa+ audio services, a sophisticated Agentic AI system that shapes the user experience for millions of users globally. This synthetic data approach enables comprehensive testing of edge cases and failure modes while protecting user privacy. Our results demonstrate that the CAT framework provides unprecedented insights into goal-task alignment, enabling more effective optimization and development of Agentic AI systems.

As machine learning models continue to grow in size and complexity, efficient serving faces increasingly broad trade-offs spanning accuracy, latency, resource usage, and other objectives. Multi-model serving further complicates these trade-offs; for example, in cascaded models, each early-exit decision balances latency reduction against potential accuracy loss. Despite the pervasiveness and importance of such trade-offs, current strategies remain largely heuristic and case-specific, limiting both their theoretical guarantees and general applicability. We present a general framework, T-Tamer, which formalizes this setting as a multi-stage decision process, where the objective is to determine both when to exit and which model to consult. Our main result shows that recall (i.e., the ability to revisit earlier models) is both necessary and sufficient for achieving provable performance guarantees. In particular, we prove that strategies without recall cannot obtain any constant-factor approximation to the optimal trade-off, whereas recall-based strategies provably attain the optimal trade-off in polynomial time. We validate our analysis through experiments on synthetic datasets and early-exit workloads for vision and NLP benchmarks. The results show that recall-based strategies consistently yield efficient accuracy-latency trade-offs. We hope this work provides a principled foundation for bridging heuristic practice with theoretical guarantees in the design of early-exit and cascaded models.

We introduce ADAM (A Diverse Archive of Mankind), a framework for evaluating and improving multimodal large language models (MLLMs) in biographical reasoning. To the best of our knowledge, this is the first work to systematically examine LLM capabilities in biography, a critical yet underexplored dimension of factual knowledge. At its core, AdamDB is a multilingual and multimodal dataset covering over 4 million individuals across geography, time, and profession, while AdamBench provides cognitively structured evaluations based on Bloom's taxonomy, spanning six reasoning levels in both English and native languages. To address hallucinations, particularly for lesser-known individuals, we propose AdamRAG, a retrieval-augmented generation system tailored to biographical contexts. Experiments show that AdamRAG substantially improves open-source models and modestly benefits closed-source ones, with the largest gains on lower-order reasoning. Popularity strongly mediates accuracy, and multimodal input via face images offers smaller, less consistent improvements than retrieval. ADAM establishes the first benchmark and framework for cognitively, culturally, and multimodally grounded biographical evaluation, advancing the development of multilingual, accurate, and hallucination-resistant MLLMs.

Large language models (LLMs) have demonstrated strong persuasive capabilities comparable to those of humans, offering promising benefits while raising societal concerns about their deployment. However, systematically evaluating the persuasive capabilities of LLMs is inherently challenging, as the effectiveness of persuasion among humans varies significantly across different domains. In this paper, we take a theory-driven approach to provide a scalable and principled framework for measuring the persuasive capabilities of LLMs. Grounded in the Bayesian Persuasion (BP) framework, we repurpose existing human-human persuasion datasets to construct environments for evaluating and training LLMs in strategic persuasion. Our results reveal that frontier models can consistently achieve high persuasion gains and exhibit sophisticated persuasion strategies that align with theoretical predictions. Building on this, we use reinforcement learning to train LLMs for strategic persuasion in our environments. Our results also demonstrate that even small LLMs can obtain significantly higher persuasion gains through reinforcement learning.

Large Language Models (LLMs) vary widely in their capabilities, with larger models often having better performance but higher cost: choosing an LLM model often involves trading off performance and cost. The LLM Cascade is a paradigm that defers difficult queries from weak/cheap to strong/expensive models. This approach is nonadaptive: the deferral decision is trained offline. When confronted with similar or repeated queries, the LLM Cascade may then repeatedly consult the expensive model and incur higher cost. To improve the cascading efficiency, we propose Inter-Cascade, an online and interactive LLM Cascade that extends the role of strong model from a backup helper to a long-term teacher. In our system, when a strong model resolves a difficult query, it also distills its solution into a generalized, reusable problem-solving strategy that boosts the weak model on subsequent queries. Adding strategies to queries enables the weak model to dynamically improve its performance over time, avoiding computationally and time-intensive fine-tuning. Empirically, compared with standard LLM Cascade baselines across multiple benchmarks, the Inter-Cascade significantly improves the accuracy of the weak model (by up to 33.06 absolute percentage points) and the overall system (by up to 5.53 absolute percentage points), while reducing the calls to strong models (by up to 48.05% relative reduction) and saving the corresponding fees (by up to 49.63% relative reduction). Inter-Cascade demonstrates the effective in-context knowledge transfer between LLMs, and provides a general, scalable framework applicable to both open-source and API-based LLMs.

As Generative AI (GenAI) systems see growing adoption, a key concern involves the external validity of evaluations, or the extent to which they generalize from lab-based to real-world deployment conditions. Threats to the external validity of GenAI evaluations arise when the source sample of human raters and system outputs used to obtain a system quality estimate differs from the target distribution at deployment time. In this work, we propose a doubly-robust estimation framework designed to address this evaluation sampling bias. Key to our approach is the use of "persona" ratings produced by prompting an LLM evaluator (i.e., an LLM-as-a-judge) to behave as a human rater with specific sociodemographic characteristics. Our doubly-robust framework combines these informative yet imperfect persona ratings with human ratings obtained under evaluation sampling bias to produce statistically valid system quality estimates. In particular, we show that our approach yields valid system quality estimates when either (i) a model trained to predict human ratings using persona ratings and source data observed under sampling bias, or (ii) a reweighting model that corrects for sampling bias is of sufficient quality. We validate our framework theoretically and via a novel Persona Simulation Framework (PSF) designed to systematically manipulate persona quality and the degree of evaluation sampling bias present in source data. Our work provides a principled foundation for combining imperfect persona ratings with human ratings observed under sampling bias to obtain valid system quality estimates.

The QMoE model provides a practical approach for compression of massive Mixture-of-Experts (MoE) models. QMoE offers a solution geared towards memory limitations that often reach terabyte scales, and it has the advantage of working with high sparsity models which implicitly lend themselves to compression techniques. QMoE also has the advantage of only taking MoE models into account and does not evaluate its use with non mixture of expert systems. Although this prior attempt focuses on the limitations of large servers with the latest NVIDIA hardware which in the case of the H100 and V100 which have 80 GB of HBM (High Bandwidth Memory), what is not being considered is a significantly more constrained environment, such as in the case of mobile devices which may have in the case of the iPhone anywhere from 4 to 8 GB of unified memory which also needs to be shared with the operating system and additional processes. Although edge devices such as phones and laptops are becoming increasingly more computationally powerful, they are still not close to the level of advanced server machines such as NVIDIA. An additional constraint that we must consider is that of latency. The communication time of sending a request to an LLM server and then getting it back is an additional waiting time that can be removed. We may also want to use LLM technology in environments where there is no reliable network connection.

The majority of deep learning-based speech enhancement methods require paired clean-noisy speech data. Collecting such data at scale in real-world conditions is infeasible, which has led the community to rely on synthetically generated noisy speech. However, this introduces a gap between the training and testing phases. In this work, we propose a novel dual-branch encoder-decoder architecture for unsupervised speech enhancement that separates the input into clean speech and residual noise. Adversarial training is employed to impose priors on each branch, defined by unpaired datasets of clean speech and, optionally, noise. Experimental results show that our method achieves performance comparable to leading unsupervised speech enhancement approaches. Furthermore, we demonstrate the critical impact of clean speech data selection on enhancement performance. In particular, our findings reveal that performance may appear overly optimistic when in-domain clean speech data are used for prior definition -- a practice adopted in previous unsupervised speech enhancement studies.