Related papers: Llasa: Scaling Train-Time and Inference-Time Compu…

Auto-regressive speech-text models pre-trained on interleaved text tokens and discretized speech tokens demonstrate strong speech understanding and generation, yet remain substantially less compute-efficient than text LLMs, partly due to…

Large language models (LLMs) have recently garnered significant interest. With in-context learning, LLMs achieve impressive results in various natural language tasks. However, the application of LLMs to sentence embeddings remains an area…

Recently, the emergence of large language models (LLMs) has motivated integrating language descriptions into graphs, forming text-attributed graphs (TAGs) that enhance model encoding capabilities from a data-centric perspective. A review of…

Recent advancements in text-to-speech (TTS) have shown that language model (LM) based systems offer competitive performance compared to traditional approaches. However, in training, TTS models use ground-truth (GT) tokens as prefixes to…

Process Reward Models (PRMs) provide step-level supervision that improves the reliability of reasoning in large language models. While PRMs have been extensively studied in text-based domains, their extension to Vision Language Models…

Automated speaking assessment (ASA) on opinion expressions is often hampered by the scarcity of labeled recordings, which restricts prompt diversity and undermines scoring reliability. To address this challenge, we propose a novel training…

Large language models are increasingly adopted as semantic backbones for neural text-to-speech systems. However, frozen LLM representations are insufficient for modeling speaker specific acoustic and perceptual characteristics. Our…

Scaling laws for large language models (LLMs) have provided useful guidance in training ever larger models for predictable performance gains. Time series forecasting shares a similar sequential structure to language, and is amenable to…

Human reasoning is shaped by resource rationality -- optimizing performance under constraints. Recently, inference-time scaling has emerged as a powerful paradigm to improve the reasoning performance of Large Language Models by expanding…

Recently, scaling test-time compute on Large Language Models (LLM) has garnered wide attention. However, there has been limited investigation of how various reasoning prompting strategies perform as scaling. In this paper, we focus on a…

Large Language Models (LLMs) demonstrate impressive capabilities, yet interaction with these models is mostly facilitated through text. Using Text-To-Speech to synthesize LLM outputs typically results in notable latency, which is…

Models like GPT-4o enable real-time interaction with large language models (LLMs) through speech, significantly enhancing user experience compared to traditional text-based interaction. However, there is still a lack of exploration on how…

Realignment becomes necessary when a language model (LM) fails to meet expected performance. We propose a flexible realignment framework that supports quantitative control of alignment degree during training and inference. This framework…

Aligning large language models (LLMs) to diverse human preferences is fundamentally challenging since criteria can often conflict with each other. Inference-time alignment methods have recently gained popularity as they allow LLMs to be…

Test-time scaling has emerged as a powerful technique for enhancing the reasoning capabilities of large language models. However, its effectiveness in medical reasoning remains uncertain, as the medical domain fundamentally differs from…

This paper introduces StyleSpeech, a novel Text-to-Speech~(TTS) system that enhances the naturalness and accuracy of synthesized speech. Building upon existing TTS technologies, StyleSpeech incorporates a unique Style Decorator structure…

Scaling laws have transformed our understanding of large language models by linking upstream metrics like cross-entropy loss to design factors such as model size, training data, and compute. However, these conventional laws fail to capture…

While large language models (LLMs) have been applied to automatic speech recognition (ASR), the task of making the model streamable remains a challenge. This paper proposes a novel model architecture, Transducer-Llama, that integrates LLMs…

Standard classification models often map inputs directly to labels without explicit reasoning, potentially limiting their performance, robustness, and interpretability. This paper introduces a novel two-stage approach to enhance text…

Drawing on constructs from psychology, prior work has identified a distinction between explicit and implicit bias in large language models (LLMs). While many LLMs undergo post-training alignment and safety procedures to avoid expressions of…