Enhancing Visual Reasoning with Autonomous Imagination in Multimodal Large Language Models

There have been recent efforts to extend the Chain-of-Thought (CoT) paradigm to Multimodal Large Language Models (MLLMs) by finding visual clues in the input scene, advancing the visual reasoning ability of MLLMs. However, current approaches are specially designed for the tasks where clue finding plays a major role in the whole reasoning process, leading to the difficulty in handling complex visual scenes where clue finding does not actually simplify the whole reasoning task. To deal with this challenge, we propose a new visual reasoning paradigm enabling MLLMs to autonomously modify the input scene to new ones based on its reasoning status, such that CoT is reformulated as conducting simple closed-loop decision-making and reasoning steps under a sequence of imagined visual scenes, leading to natural and general CoT construction. To implement this paradigm, we introduce a novel plug-and-play imagination space, where MLLMs conduct visual modifications through operations like focus, ignore, and transform based on their native reasoning ability without specific training. We validate our approach through a benchmark spanning dense counting, simple jigsaw puzzle solving, and object placement, challenging the reasoning ability beyond clue finding. The results verify that while existing techniques fall short, our approach enables MLLMs to effectively reason step by step through autonomous imagination. Project page: https://future-item.github.io/autoimagine-site.

Beyond Simple Sum of Delayed Rewards: Non-Markovian Reward Modeling for Reinforcement Learning

Reinforcement Learning (RL) empowers agents to acquire various skills by learning from reward signals. Unfortunately, designing high-quality instance-level rewards often demands significant effort. An emerging alternative, RL with delayed reward, focuses on learning from rewards presented periodically, which can be obtained from human evaluators assessing the agent's performance over sequences of behaviors. However, traditional methods in this domain assume the existence of underlying Markovian rewards and that the observed delayed reward is simply the sum of instance-level rewards, both of which often do not align well with real-world scenarios. In this paper, we introduce the problem of RL from Composite Delayed Reward (RLCoDe), which generalizes traditional RL from delayed rewards by eliminating the strong assumption. We suggest that the delayed reward may arise from a more complex structure reflecting the overall contribution of the sequence. To address this problem, we present a framework for modeling composite delayed rewards, using a weighted sum of non-Markovian components to capture the different contributions of individual steps. Building on this framework, we propose Composite Delayed Reward Transformer (CoDeTr), which incorporates a specialized in-sequence attention mechanism to effectively model these contributions. We conduct experiments on challenging locomotion tasks where the agent receives delayed rewards computed from composite functions of observable step rewards. The experimental results indicate that CoDeTr consistently outperforms baseline methods across evaluated metrics. Additionally, we demonstrate that it effectively identifies the most significant time steps within the sequence and accurately predicts rewards that closely reflect the environment feedback.

Enabling Synergistic Full-Body Control in Prompt-Based Co-Speech Motion Generation

Current co-speech motion generation approaches usually focus on upper body gestures following speech contents only, while lacking supporting the elaborate control of synergistic full-body motion based on text prompts, such as talking while walking. The major challenges lie in 1) the existing speech-to-motion datasets only involve highly limited full-body motions, making a wide range of common human activities out of training distribution; 2) these datasets also lack annotated user prompts. To address these challenges, we propose SynTalker, which utilizes the off-the-shelf text-to-motion dataset as an auxiliary for supplementing the missing full-body motion and prompts. The core technical contributions are two-fold. One is the multi-stage training process which obtains an aligned embedding space of motion, speech, and prompts despite the significant distributional mismatch in motion between speech-to-motion and text-to-motion datasets. Another is the diffusion-based conditional inference process, which utilizes the separate-then-combine strategy to realize fine-grained control of local body parts. Extensive experiments are conducted to verify that our approach supports precise and flexible control of synergistic full-body motion generation based on both speeches and user prompts, which is beyond the ability of existing approaches.

Reinforcement Learning from Bagged Reward: A Transformer-based Approach for Instance-Level Reward Redistribution

In reinforcement Learning (RL), an instant reward signal is generated for each action of the agent, such that the agent learns to maximize the cumulative reward to obtain the optimal policy. However, in many real-world applications, the instant reward signals are not obtainable by the agent. Instead, the learner only obtains rewards at the ends of bags, where a bag is defined as a partial sequence of a complete trajectory. In this situation, the learner has to face the significant difficulty of exploring the unknown instant rewards in the bags, which could not be addressed by existing approaches, including those trajectory-based approaches that consider only complete trajectories and ignore the inner reward distributions. To formally study this situation, we introduce a novel RL setting termed Reinforcement Learning from Bagged Rewards (RLBR), where only the bagged rewards of sequences can be obtained. We provide the theoretical study to establish the connection between RLBR and standard RL in Markov Decision Processes (MDPs). To effectively explore the reward distributions within the bagged rewards, we propose a Transformer-based reward model, the Reward Bag Transformer (RBT), which uses the self-attention mechanism for interpreting the contextual nuances and temporal dependencies within each bag. Extensive experimental analyses demonstrate the superiority of our method, particularly in its ability to mimic the original MDP's reward distribution, highlighting its proficiency in contextual understanding and adaptability to environmental dynamics.

Generating by Understanding: Neural Visual Generation with Logical Symbol Groundings

Despite the great success of neural visual generative models in recent years, integrating them with strong symbolic knowledge reasoning systems remains a challenging task. The main challenges are two-fold: one is symbol assignment, i.e. bonding latent factors of neural visual generators with meaningful symbols from knowledge reasoning systems. Another is rule learning, i.e. learning new rules, which govern the generative process of the data, to augment the knowledge reasoning systems. To deal with these symbol grounding problems, we propose a neural-symbolic learning approach, Abductive Visual Generation (AbdGen), for integrating logic programming systems with neural visual generative models based on the abductive learning framework. To achieve reliable and efficient symbol assignment, the quantized abduction method is introduced for generating abduction proposals by the nearest-neighbor lookups within semantic codebooks. To achieve precise rule learning, the contrastive meta-abduction method is proposed to eliminate wrong rules with positive cases and avoid less-informative rules with negative cases simultaneously. Experimental results on various benchmark datasets show that compared to the baselines, AbdGen requires significantly fewer instance-level labeling information for symbol assignment. Furthermore, our approach can effectively learn underlying logical generative rules from data, which is out of the capability of existing approaches.

Model Spider: Learning to Rank Pre-Trained Models Efficiently

Figuring out which Pre-Trained Model (PTM) from a model zoo fits the target task is essential to take advantage of plentiful model resources. With the availability of numerous heterogeneous PTMs from diverse fields, efficiently selecting the most suitable PTM is challenging due to the time-consuming costs of carrying out forward or backward passes over all PTMs. In this paper, we propose Model Spider, which tokenizes both PTMs and tasks by summarizing their characteristics into vectors to enable efficient PTM selection. By leveraging the approximated performance of PTMs on a separate set of training tasks, Model Spider learns to construct tokens and measure the fitness score between a model-task pair via their tokens. The ability to rank relevant PTMs higher than others generalizes to new tasks. With the top-ranked PTM candidates, we further learn to enrich task tokens with their PTM-specific semantics to re-rank the PTMs for better selection. Model Spider balances efficiency and selection ability, making PTM selection like a spider preying on a web. Model Spider demonstrates promising performance in various configurations of model zoos.

Seeing Differently, Acting Similarly: Imitation Learning with Heterogeneous Observations

In many real-world imitation learning tasks, the demonstrator and the learner have to act in different but full observation spaces. This situation generates significant obstacles for existing imitation learning approaches to work, even when they are combined with traditional space adaptation techniques. The main challenge lies in bridging expert's occupancy measures to learner's dynamically changing occupancy measures under the different observation spaces. In this work, we model the above learning problem as Heterogeneous Observations Imitation Learning (HOIL). We propose the Importance Weighting with REjection (IWRE) algorithm based on the techniques of importance-weighting, learning with rejection, and active querying to solve the key challenge of occupancy measure matching. Experimental results show that IWRE can successfully solve HOIL tasks, including the challenging task of transforming the vision-based demonstrations to random access memory (RAM)-based policies under the Atari domain.

Expert-Level Atari Imitation Learning from Demonstrations Only

One of the key issues for imitation learning lies in making policy learned from limited samples to generalize well in the whole state-action space. This problem is much more severe in high-dimensional state environments, such as game playing with raw pixel inputs. Under this situation, even state-of-the-art adversary based imitation learning algorithms fail. Through theoretical and empirical studies, we find that the main cause lies in the failure of training a powerful discriminator to generate meaningful rewards in high-dimensional environments. Theoretical results are provided to suggest the necessity of dimensionality reduction. However, since preserving important discriminative information via feature transformation is a non-trivial task, a straightforward application of off-the-shelf methods cannot achieve desirable performance. To address the above issues, we propose HashReward, which is a novel imitation learning algorithm utilizing the idea of supervised hashing to realize effective training of the discriminator. As far as we are aware, HashReward is the first pure imitation learning approach to achieve expert comparable performance in Atari game environments with raw pixel inputs.

Crowdsourcing with Unsure Option

One of the fundamental problems in crowdsourcing is the trade-off between the number of the workers needed for high-accuracy aggregation and the budget to pay. For saving budget, it is important to ensure high quality of the crowd-sourced labels, hence the total cost on label collection will be reduced. Since the self-confidence of the workers often has a close relationship with their abilities, a possible way for quality control is to request the workers to return the labels only when they feel confident, by means of providing unsure option to them. On the other hand, allowing workers to choose unsure option also leads to the potential danger of budget waste. In this work, we propose the analysis towards understanding when providing the unsure option indeed leads to significant cost reduction, as well as how the confidence threshold is set. We also propose an online mechanism, which is alternative for threshold selection when the estimation of the crowd ability distribution is difficult.