Self-Consistency Preference Optimization

Self-alignment, whereby models learn to improve themselves without human annotation, is a rapidly growing research area. However, existing techniques often fail to improve complex reasoning tasks due to the difficulty of assigning correct rewards. An orthogonal approach that is known to improve correctness is self-consistency, a method applied at inference time based on multiple sampling in order to find the most consistent answer. In this work, we extend the self-consistency concept to help train models. We thus introduce self-consistency preference optimization (ScPO), which iteratively trains consistent answers to be preferred over inconsistent ones on unsupervised new problems. We show ScPO leads to large improvements over conventional reward model training on reasoning tasks such as GSM8K and MATH, closing the gap with supervised training with gold answers or preferences, and that combining ScPO with standard supervised learning improves results even further. On ZebraLogic, ScPO finetunes Llama-3 8B to be superior to Llama-3 70B, Gemma-2 27B, and Claude-3 Haiku.

EMOVA: Empowering Language Models to See, Hear and Speak with Vivid Emotions

GPT-4o, an omni-modal model that enables vocal conversations with diverse emotions and tones, marks a milestone for omni-modal foundation models. However, empowering Large Language Models to perceive and generate images, texts, and speeches end-to-end with publicly available data remains challenging in the open-source community. Existing vision-language models rely on external tools for the speech processing, while speech-language models still suffer from limited or even without vision-understanding abilities. To address this gap, we propose EMOVA (EMotionally Omni-present Voice Assistant), to enable Large Language Models with end-to-end speech capabilities while maintaining the leading vision-language performance. With a semantic-acoustic disentangled speech tokenizer, we notice surprisingly that omni-modal alignment can further enhance vision-language and speech abilities compared with the corresponding bi-modal aligned counterparts. Moreover, a lightweight style module is proposed for flexible speech style controls (e.g., emotions and pitches). For the first time, EMOVA achieves state-of-the-art performance on both the vision-language and speech benchmarks, and meanwhile, supporting omni-modal spoken dialogue with vivid emotions.

PSFHS Challenge Report: Pubic Symphysis and Fetal Head Segmentation from Intrapartum Ultrasound Images

Segmentation of the fetal and maternal structures, particularly intrapartum ultrasound imaging as advocated by the International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) for monitoring labor progression, is a crucial first step for quantitative diagnosis and clinical decision-making. This requires specialized analysis by obstetrics professionals, in a task that i) is highly time- and cost-consuming and ii) often yields inconsistent results. The utility of automatic segmentation algorithms for biometry has been proven, though existing results remain suboptimal. To push forward advancements in this area, the Grand Challenge on Pubic Symphysis-Fetal Head Segmentation (PSFHS) was held alongside the 26th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2023). This challenge aimed to enhance the development of automatic segmentation algorithms at an international scale, providing the largest dataset to date with 5,101 intrapartum ultrasound images collected from two ultrasound machines across three hospitals from two institutions. The scientific community's enthusiastic participation led to the selection of the top 8 out of 179 entries from 193 registrants in the initial phase to proceed to the competition's second stage. These algorithms have elevated the state-of-the-art in automatic PSFHS from intrapartum ultrasound images. A thorough analysis of the results pinpointed ongoing challenges in the field and outlined recommendations for future work. The top solutions and the complete dataset remain publicly available, fostering further advancements in automatic segmentation and biometry for intrapartum ultrasound imaging.

Enhancing Multilingual Speech Generation and Recognition Abilities in LLMs with Constructed Code-switched Data

While large language models (LLMs) have been explored in the speech domain for both generation and recognition tasks, their applications are predominantly confined to the monolingual scenario, with limited exploration in multilingual and code-switched (CS) contexts. Additionally, speech generation and recognition tasks are often handled separately, such as VALL-E and Qwen-Audio. In this paper, we propose a MutltiLingual MultiTask (MLMT) model, integrating multilingual speech generation and recognition tasks within the single LLM. Furthermore, we develop an effective data construction approach that splits and concatenates words from different languages to equip LLMs with CS synthesis ability without relying on CS data. The experimental results demonstrate that our model outperforms other baselines with a comparable data scale. Furthermore, our data construction approach not only equips LLMs with CS speech synthesis capability with comparable speaker consistency and similarity to any given speaker, but also improves the performance of LLMs in multilingual speech generation and recognition tasks.

DexSim2Real$^{2}$: Building Explicit World Model for Precise Articulated Object Dexterous Manipulation

Articulated object manipulation is ubiquitous in daily life. In this paper, we present DexSim2Real$^{2}$, a novel robot learning framework for goal-conditioned articulated object manipulation using both two-finger grippers and multi-finger dexterous hands. The key of our framework is constructing an explicit world model of unseen articulated objects through active one-step interactions. This explicit world model enables sampling-based model predictive control to plan trajectories achieving different manipulation goals without needing human demonstrations or reinforcement learning. It first predicts an interaction motion using an affordance estimation network trained on self-supervised interaction data or videos of human manipulation from the internet. After executing this interaction on the real robot, the framework constructs a digital twin of the articulated object in simulation based on the two point clouds before and after the interaction. For dexterous multi-finger manipulation, we propose to utilize eigengrasp to reduce the high-dimensional action space, enabling more efficient trajectory searching. Extensive experiments validate the framework's effectiveness for precise articulated object manipulation in both simulation and the real world using a two-finger gripper and a 16-DoF dexterous hand. The robust generalizability of the explicit world model also enables advanced manipulation strategies, such as manipulating with different tools.

Functionally Constrained Algorithm Solves Convex Simple Bilevel Problems

This paper studies simple bilevel problems, where a convex upper-level function is minimized over the optimal solutions of a convex lower-level problem. We first show the fundamental difficulty of simple bilevel problems, that the approximate optimal value of such problems is not obtainable by first-order zero-respecting algorithms. Then we follow recent works to pursue the weak approximate solutions. For this goal, we propose novel near-optimal methods for smooth and nonsmooth problems by reformulating them into functionally constrained problems.

ActiveNeRF: Learning Accurate 3D Geometry by Active Pattern Projection

NeRFs have achieved incredible success in novel view synthesis. However, the accuracy of the implicit geometry is unsatisfactory because the passive static environmental illumination has low spatial frequency and cannot provide enough information for accurate geometry reconstruction. In this work, we propose ActiveNeRF, a 3D geometry reconstruction framework, which improves the geometry quality of NeRF by actively projecting patterns of high spatial frequency onto the scene using a projector which has a constant relative pose to the camera. We design a learnable active pattern rendering pipeline which jointly learns the scene geometry and the active pattern. We find that, by adding the active pattern and imposing its consistency across different views, our proposed method outperforms state of the art geometry reconstruction methods qualitatively and quantitatively in both simulation and real experiments. Code is avaliable at https://github.com/hcp16/active_nerf

Con4m: Context-aware Consistency Learning Framework for Segmented Time Series Classification

Time Series Classification (TSC) encompasses two settings: classifying entire sequences or classifying segmented subsequences. The raw time series for segmented TSC usually contain Multiple classes with Varying Duration of each class (MVD). Therefore, the characteristics of MVD pose unique challenges for segmented TSC, yet have been largely overlooked by existing works. Specifically, there exists a natural temporal dependency between consecutive instances (segments) to be classified within MVD. However, mainstream TSC models rely on the assumption of independent and identically distributed (i.i.d.), focusing on independently modeling each segment. Additionally, annotators with varying expertise may provide inconsistent boundary labels, leading to unstable performance of noise-free TSC models. To address these challenges, we first formally demonstrate that valuable contextual information enhances the discriminative power of classification instances. Leveraging the contextual priors of MVD at both the data and label levels, we propose a novel consistency learning framework Con4m, which effectively utilizes contextual information more conducive to discriminating consecutive segments in segmented TSC tasks, while harmonizing inconsistent boundary labels for training. Extensive experiments across multiple datasets validate the effectiveness of Con4m in handling segmented TSC tasks on MVD.

Meta-Rewarding Language Models: Self-Improving Alignment with LLM-as-a-Meta-Judge

Large Language Models (LLMs) are rapidly surpassing human knowledge in many domains. While improving these models traditionally relies on costly human data, recent self-rewarding mechanisms (Yuan et al., 2024) have shown that LLMs can improve by judging their own responses instead of relying on human labelers. However, existing methods have primarily focused on improving model responses rather than judgment capabilities, resulting in rapid saturation during iterative training. To address this issue, we introduce a novel Meta-Rewarding step to the self-improvement process, where the model judges its own judgements and uses that feedback to refine its judgment skills. Surprisingly, this unsupervised approach improves the model's ability to judge {\em and} follow instructions, as demonstrated by a win rate improvement of Llama-3-8B-Instruct from 22.9% to 39.4% on AlpacaEval 2, and 20.6% to 29.1% on Arena-Hard. These results strongly suggest the potential for self-improving models without human supervision.

Dilated Strip Attention Network for Image Restoration

Image restoration is a long-standing task that seeks to recover the latent sharp image from its deteriorated counterpart. Due to the robust capacity of self-attention to capture long-range dependencies, transformer-based methods or some attention-based convolutional neural networks have demonstrated promising results on many image restoration tasks in recent years. However, existing attention modules encounters limited receptive fields or abundant parameters. In order to integrate contextual information more effectively and efficiently, in this paper, we propose a dilated strip attention network (DSAN) for image restoration. Specifically, to gather more contextual information for each pixel from its neighboring pixels in the same row or column, a dilated strip attention (DSA) mechanism is elaborately proposed. By employing the DSA operation horizontally and vertically, each location can harvest the contextual information from a much wider region. In addition, we utilize multi-scale receptive fields across different feature groups in DSA to improve representation learning. Extensive experiments show that our DSAN outperforms state-of-the-art algorithms on several image restoration tasks.