Search Results for author:
Found 22 papers, 5 papers with code
Towards Better Generalization via Distributional Input Projection Network
As overparameterized models become increasingly prevalent, training loss alone offers limited insight into generalization performance.
QiMeng: Fully Automated Hardware and Software Design for Processor Chip
In the middle-layer, leveraging the LPCM's knowledge representation and inference capabilities, we develop the Hardware Design Agent and the Software Design Agent to automate the design of hardware and software for processor chips.
Optimizing Chain-of-Thought Reasoners via Gradient Variance Minimization in Rejection Sampling and RL
Chain-of-thought (CoT) reasoning in large language models (LLMs) can be formalized as a latent variable problem, where the model needs to generate intermediate reasoning steps.
QiMeng-Xpiler: Transcompiling Tensor Programs for Deep Learning Systems with a Neural-Symbolic Approach
We propose a novel transcompiler, i. e., QiMeng-Xpiler, for automatically translating tensor programs across DLS via both large language models (LLMs) and symbolic program synthesis, i. e., neural-symbolic synthesis.
Object-Level Verbalized Confidence Calibration in Vision-Language Models via Semantic Perturbation
In this work, we propose a novel Confidence Calibration through Semantic Perturbation (CSP) framework to improve the calibration of verbalized confidence for VLMs in response to object-centric queries.
Efficient Model Editing with Task Vector Bases: A Theoretical Framework and Scalable Approach
Task vectors, which are derived from the difference between pre-trained and fine-tuned model weights, enable flexible task adaptation and model merging through arithmetic operations such as addition and negation.
SEEN-DA: SEmantic ENtropy guided Domain-aware Attention for Domain Adaptive Object Detection
Domain adaptive object detection (DAOD) aims to generalize detectors trained on an annotated source domain to an unlabelled target domain.
DA-Ada: Learning Domain-Aware Adapter for Domain Adaptive Object Detection
Domain adaptive object detection (DAOD) aims to generalize detectors trained on an annotated source domain to an unlabelled target domain.
SPDiffusion: Semantic Protection Diffusion for Multi-concept Text-to-image Generation
We found that attribute confusion occurs when a certain region of the latent features attend to multiple or incorrect prompt tokens.
Towards Analyzing and Mitigating Sycophancy in Large Vision-Language Models
For improvement, we propose Leading Query Contrastive Decoding (LQCD), a model-agnostic method focusing on calibrating the LVLMs' over-reliance on leading cues by identifying and suppressing the probabilities of sycophancy tokens at the decoding stage.
HeteroMorpheus: Universal Control Based on Morphological Heterogeneity Modeling
In the field of robotic control, designing individual controllers for each robot leads to high computational costs.
TensorTEE: Unifying Heterogeneous TEE Granularity for Efficient Secure Collaborative Tensor Computing
However, existing heterogeneous TEE designs are inefficient for collaborative computing due to fine and different memory granularities between CPU and NPU.
Luban: Building Open-Ended Creative Agents via Autonomous Embodied Verification
In this work, we introduce autonomous embodied verification techniques for agents to fill the gap, laying the groundwork for creative tasks.
On the Benefits of Over-parameterization for Out-of-Distribution Generalization
We demonstrate that in this scenario, further increasing the model's parameterization can significantly reduce the OOD loss.
Assessing and Understanding Creativity in Large Language Models
This paper aims to establish an efficient framework for assessing the level of creativity in LLMs.
The Surprising Harmfulness of Benign Overfitting for Adversarial Robustness
Recent empirical and theoretical studies have established the generalization capabilities of large machine learning models that are trained to (approximately or exactly) fit noisy data.
Spurious Feature Diversification Improves Out-of-distribution Generalization
Contrary to the conventional wisdom that focuses on learning invariant features for better OOD performance, our findings suggest that incorporating a large number of diverse spurious features weakens their individual contributions, leading to improved overall OOD generalization performance.
Pushing the Limits of Machine Design: Automated CPU Design with AI
By efficiently exploring a search space of unprecedented size 10^{10^{540}}, which is the largest one of all machine-designed objects to our best knowledge, and thus pushing the limits of machine design, our approach generates an industrial-scale RISC-V CPU within only 5 hours.
Ultra-low Precision Multiplication-free Training for Deep Neural Networks
The energy-efficient works try to decrease the precision of multiplication or replace the multiplication with energy-efficient operations such as addition or bitwise shift, to reduce the energy consumption of FP32 multiplications.
Class-Specific Attention (CSA) for Time-Series Classification
In this paper, we propose a novel class-specific attention (CSA) module to capture significant class-specific features and improve the overall classification performance of time series.
Scheduling Real-time Deep Learning Services as Imprecise Computations
The paper presents an efficient real-time scheduling algorithm for intelligent real-time edge services, defined as those that perform machine intelligence tasks, such as voice recognition, LIDAR processing, or machine vision, on behalf of local embedded devices that are themselves unable to support extensive computations.
