All papers

The paper introduces BanglaTOD, a tri-lingual benchmark for task-oriented dialogue understanding covering Bangla, English, and romanized Banglish across nine service domains.

The paper introduces an optimization method that jointly refines communication instructions between communicating agents to improve multi-agent collaboration.

The paper probes language model hidden states across generated stories to test whether models anticipate narrative events multiple steps ahead.

The paper presents SparPR, a runtime key-value cache recovery mechanism that reduces memory bottlenecks during model inference on consumer hardware.

Presents a pragmatic prompting framework to preserve speaker voice, emotion, and tone during speech-to-speech translation.

Surveys methods for compressing large language model benchmarks to reduce evaluation costs while maintaining accurate score estimates.

Presents a multilingual framework and dataset spanning four languages to evaluate clinical pain representations across exams and clinical notes.

Introduces a human-verified benchmark to evaluate how accurately reward models judge language model outputs on legal tasks.

Develops a simulation framework using persona-guided models to optimize comparative doctor summary presentations for medical patients.

Probes layer representations in multilingual language models to study whether internal color geometry aligns with human perception or language syntax.

Equips diffusion language models with multiple residual streams and adaptive routing to separate visible context from masked token generation.

Proposes estimating both contextual and unigram surprisal from the same language model to reduce estimation mismatch in reading time studies.

Estimates surprisal directly from continuous speech signals using a speech language model and importance sampling over abstract linguistic units.

Proposes a teacher-guided reinforcement learning method using bidirectional constraints to help small language models learn effective tool use.

Introduces a benchmark evaluating how reliably software coding agents extract and digitize complex unstructured clinical trial protocol documents.

Analyzes retrieval divergence across languages in multilingual dense embedders and introduces disagreement-guided steering to align shared-index search results.

This paper introduces a visual associative test benchmark to evaluate the lateral reasoning and creative association capabilities of vision-language models.

This paper introduces a multi-agent retrieval framework that coordinates search across legal statutes and precedent cases for grounded legal reasoning.

This paper analyzes multi-person conversational involvement by jointly evaluating speaking floors and spatial formations from multimodal video features.

This paper evaluates pretraining on chemical language models across 64 molecular datasets, finding that gains saturate below gradient-boosted tree baselines.

This paper proposes a Bayesian method that infers task-aligned similarity metrics at test time using weak interactive feedback.

This paper presents a method to detect spoken code-switching directly from speech representations using language-selective hidden dimensions.

This paper introduces a reinforcement learning scaffold that selectively injects textual skills into agents using low-cost utility estimation.

This paper introduces a benchmark to evaluate multimodal models on joint audio-visual temporal grounding, spatial grounding, and question answering.

This paper frames Earth observation model development as an automated loop where coding agents build and iteratively refine models using verifiable harnesses.

This paper presents a robotic manipulation framework combining high-level language planning with vision-language verification to enforce physical and kinematic constraints.

This paper introduces a training framework that enables harness-based software agents to self-improve without relying on external teacher models.

This paper introduces a learned correction step to reduce error accumulation and the training-inference gap in iterative diffusion model sampling.

This paper updates a long-horizon terminal benchmark to measure sustained agent progress across complex engineering workflows using dense process rewards.

This paper presents an ecosystem of benchmark datasets and evaluation frameworks to advance machine learning applications in veterinary medicine.

This paper evaluates whether combining radiomics with deep image representations improves probability calibration and discrimination in renal cancer classification models.

This paper analyzes decision logits in radiomics-image fusion models to distinguish gate weight values from actual feature contributions during cancer classification.

This paper introduces a coarse-to-fine text generation method using discrete code hierarchies to accelerate parallel sampling over diffusion language models.

This paper presents a synthetic multi-panel chart visual question answering dataset to evaluate how multimodal models interpret complex scientific figures.

This paper evaluates open-weight language models against closed frontier models on a cybersecurity detection benchmark using an interactive telemetry environment.

The paper uses counterfactual interventions to visualize and examine implicit environmental concepts learned inside a reinforcement learning agent's value function.

The authors introduce sparse autoencoders to trace multimodal training influence down to specific image regions and text phrases.

The paper proposes a learnable low-rank polynomial sketch to reduce approximation errors in linear attention mechanisms.

The authors propose a two-level recursive self-improvement framework to evolve and refine execution harnesses for open-world agents.

The paper develops and evaluates a multimodal foundation model for interpreting oncologic chest, abdomen, and pelvis CT scans.

The paper presents a latent world action model that predicts action-induced state transitions rather than full-screen pixel reconstructions for graphical agents.

The authors present a framework where autonomous agents iteratively refine data engineering strategies to improve model training outcomes.

The paper introduces a sparse interaction diffusion method that coordinates multi-robot trajectories using a frozen single-robot prior and modular corrections.

The authors develop an error-compensated 4-bit all-reduce algorithm to reduce communication overhead during distributed model training and inference.

The paper studies gate learning in recurrent networks, showing how adding short-dependency samples helps models learn long-range temporal connections.

The authors introduce a framework that uses cross-view depth consistency as a single reward to improve 4D video generation geometry.

The paper proposes a training-free method leveraging pullback directions to increase sample diversity in conditional diffusion generation.

The authors propose decoupled positional encodings enabling full-duplex speech models to run internal thinking streams at hardware-adaptive speeds.

The paper introduces a receding-horizon optimal control algorithm for constrained sampling in flow matching generative models.

The authors present a feed-forward network that aggregates local wall predictions to reconstruct floor plans from sparse unposed panoramas.

The paper presents an elastic vision architecture that dynamically reconfigures width and supports partitionable inference across heterogeneous edge devices.

The paper introduces a post-training residual correction method to predict and correct endpoint errors during low-step generative model sampling.

The authors run controlled experiments showing how parameter schema differences in metadata alter language model tool selection.

The authors present a foundation controller that converts motion drafts into contact-consistent joint commands for dexterous robotic hands.

The paper introduces a neuro-symbolic benchmark to evaluate whether language model agents actively investigate environments and make causal decisions.

The paper reviews deep learning architectures and training dynamics through the lens of continuous-time nonlinear control theory.

The paper measures inference energy efficiency across 86 open-weight language models, showing large efficiency gaps among models of similar capability.

The paper presents a simulated benchmark across eleven unconstrained environments to evaluate vision-language models on documenting real-world human physical work.

The paper presents a reinforcement learning framework that trains language models to solve complex logical reasoning tasks using interactive symbolic solvers.

The paper proposes an uncurated multi-stream latent memory framework that augments vision-language-action robotic policies while maintaining low inference latency.