The rising popularity of large foundation models has led to a heightened demand for parameter-efficient fine-tuning methods, such as Low-Rank Adaptation (LoRA), which offer performance comparable to full model fine-tuning while requiring only a few additional parameters tailored to the specific base model. When such base models are deprecated and replaced, all associated LoRA modules must be retrained, requiring access to either the original training data or a substantial amount of synthetic data that mirrors the original distribution. However, the original data is often inaccessible due to privacy or licensing issues, and generating synthetic data may be impractical and insufficiently representative. These factors complicate the fine-tuning process considerably. To address this challenge, we introduce a new adapter, Cross-Model Low-Rank Adaptation (LoRA-X), which enables the training-free transfer of LoRA parameters across source and target models, eliminating the need for original or synthetic training data. Our approach imposes the adapter to operate within the subspace of the source base model. This constraint is necessary because our prior knowledge of the target model is limited to its weights, and the criteria for ensuring the adapter’s transferability are restricted to the target base model’s weights and subspace. To facilitate the transfer of LoRA parameters of the source model to a target model, we employ the adapter only in the layers of the target model that exhibit an acceptable level of subspace similarity. Our extensive experiments demonstrate the effectiveness of LoRA-X for text-to-image generation, including Stable Diffusion v1.5 and Stable Diffusion XL.

Recently, 3D Gaussian Splatting (3DGS) has emerged as a significant advancement in 3D scene reconstruction, attracting considerable attention due to its ability to recover high-fidelity details while maintaining low complexity. Despite the promising results achieved by 3DGS, its rendering performance is constrained by its dependence on costly non-commutative alpha-blending operations. These operations mandate complex view dependent sorting operations that introduce computational overhead, especially on the resource-constrained platforms such as mobile phones. In this paper, we propose Weighted Sum Rendering, which approximates alpha blending with weighted sums, thereby removing the need for sorting. This simplifies implementation, delivers superior performance, and eliminates the ``popping'' artifacts caused by sorting. Experimental results show that optimizing a generalized Gaussian splatting formulation to the new differentiable rendering yields competitive image quality. The method was implemented and tested in a mobile device GPU, achieving on average $1.23\times$ faster rendering.

We introduce ProLoRA, enabling zero-shot adaptation of parameter-efficient fine-tuning in text-to-image diffusion models. ProLoRA transfers pre-trained low-rank adjustments (e.g., LoRA) from a source to a target model without additional training data. This overcomes the limitations of traditional methods that require retraining when switching base models, often challenging due to data constraints. ProLoRA achieves this via projection of source adjustments into the target model’s weight space, leveraging subspace and null space similarities and selectively targeting aligned layers. Evaluations on established text-to-image models demonstrate successful knowledge transfer and comparable performance without retraining.

The rise of new video modalities like virtual reality or autonomous driving has increased the demand for efficient multi-view video compression methods, both in terms of rate-distortion (R-D) performance and in terms of delay and runtime. While most recent stereo video compression approaches have shown promising performance, they compress left and right views sequentially, leading to poor parallelization and runtime performance. This work presents Low-Latency neural codec for Stereo video Streaming (LLSS), a novel parallel stereo video coding method designed for fast and efficient low-latency stereo video streaming. Instead of using a sequential cross-view motion compensation like existing methods, LLSS introduces a bidirectional feature shifting module to directly exploit mutual information among views and encode them effectively with a joint cross-view prior model for entropy coding. Thanks to this design, LLSS processes left and right views in parallel, minimizing latency; all while substantially improving R-D performance compared to both existing neural and conventional codecs.

Video compression systems must support increasing bandwidth and data throughput at low cost and power, and can be limited by entropy coding bottlenecks. Efficiency can be greatly improved by parallelizing coding, which can be done at much larger scales with new neural-based codecs, but with some compression loss related to data organization. We analyze the bit rate overhead needed to support multiple bitstreams for concurrent decoding, and for its minimization propose a method for compressing parallel-decoding entry points, using bidirectional bitstream packing, and a new form of jointly optimizing arithmetic coding termination. It is shown that those techniques significantly lower the overhead, making it easier to reduce it to a small fraction of the average bitstream size, like, for example, less than 1% and 0.1% when the average number of bitstream bytes is respectively larger than 95 and 1,200 bytes.

In an information-theoretic framework, we introduce a two-stage decoding scheme capable of achieving identification capacity to address search and memory complexities in large-scale identification systems. This two-stage decoding procedure is accomplished as follows. For a given query, at the first stage, a list of cluster indices is estimated. Then, at the second stage, refinement checks are performed for all the members of the clusters to produce a single index. The first result this paper presents is the achievable rate quadruple region that specifies necessary conditions that the two-stage decoding scheme should satisfy to be able to achieve the identification capacity. The rest of this paper is designated to investigate various achievable rate quadruples in which the proposed two-stage identification setup can reduce the search complexity with respect to conventional identification setups.

We introduce a scheme to address the trade-off between the identification rate, search and memory complexities in large-scale identification systems. We use a special database organization by assigning database entries to a set of possibly overlapping clusters. The clusters are generated based on statistics of both database entries and queries. The decoding procedure is accomplished in two stages. First, a list of clusters related to the query is detected. Then, refinement checks are performed on members of the detected clusters to produce a unique index. We investigate the minimum achievable search complexity for binary symmetric sources.

Content fingerprinting and digital watermarking are techniques that are used for content protection and distribution monitoring and, more recently, for interaction with physical objects. Over the past few years, both techniques have been well studied and their shortcomings understood. In this paper, we introduce a new framework called active content fingerprinting, which takes the best from two worlds of content fingerprinting and digital watermarking, in order to overcome some of the fundamental restrictions of these techniques in terms of performance and complexity. The proposed framework extends the encoding process of conventional content fingerprinting in a way similar to digital watermarking, thus allowing the extraction of fingerprints from the modified cover data. We consider several encoding strategies, examine the performance of the proposed schemes in terms of bit error rate, the probabilities of correct identification and false acceptance and compare it with those of conventional fingerprinting and digital watermarking. Finally, we extend the proposed framework to the multidimensional case based on lattices and demonstrate its performance on both synthetic data and real images.

In this paper, we extend a new framework introduced as active content fingerprinting in [1] 1 that takes the best from the two worlds of content fingerprinting and digital watermarking to overcome some of the fundamental restrictions of these techniques in terms of performance and complexity. In the proposed framework, contents are modified in a way similar to watermarking to extract more robust fingerprints in contrast to conventional content fingerprinting. We investigate the performance of two modulation techniques based on unidimensional shrinkage and multidimensional lattice quantization. The simulation results on real images demonstrate the high efficiency of the proposed methods facing low-quality compression and additive noise.

In this paper, we introduce a new generalized scheme to resolve the trade-off between the identification rate, search and memory complexities in large-scale identification systems. The main contribution of this paper consists in a special database organization based on assigning entries of a database to a set of predefined and possibly overlapping clusters, where the cluster representative points are generated based on statistics of both entries of the database and queries. The decoding procedure is accomplished in two stages: At the first stage, a list of clusters related to the query is estimated, then refinement checks are performed to all members of these clusters to produce a unique index at the second stage. The proposed scheme generalizes several practical searching in identification systems as well as makes it possible to approach a new achievable region of search- memory complexity trade-off.

This paper is dedicated to the performance analysis of content-based identification using binary fingerprints and constrained list-based decoding. We formulate content-based identification as a multiple hypothesis test and develop analytical models of its performance in terms of probabilities of correct detection/miss and false acceptance for a class of statistical models, which captures the correlation between elements of either the content or its extracted features. Furthermore, in order to determine the block/codeword length impact on the identification's accuracy, we analyze exponents of these probabilities of errors. Finally, we develop a probabilistic model, justifying the accuracy of identification based on list decoding by evaluating the position of the queried entry on the output list. The obtained results make it possible to characterize the performance of traditional unique decoding, based on the maximum likelihood for the situations when the decoder fails to produce the correct index. This paper also contains experimental results that confirm theoretical findings.

This paper is dedicated to the analysis of object identification under desynchronization distortions. While this class of degradations is almost unavoidable in the functionality of such systems especially at the verification stage via acquisition imperfections, currently available their performance limits do not take its impact into consideration. In this paper we will try to close this gap providing the estimation of the achievable identification rates due to desynchronization distortions. Finally, the impact of the codeword length on the identification performance is justified.

A number of different multimedia fingerprinting algorithms and identification techniques were proposed and analyzed recently. This paper presents a content identification setup for a class of multimedia data that can be modeled by the Gauss-Markov process. We advocate a constrained order statistics decoding scheme based on digital fingerprints extracted from correlated data to identify contents. Finally, we investigate the fundamental limits of the proposed setup by deriving bounds on the miss and false acceptance probabilities.

In this paper we analyze the problem of object identification in channels with desynchronization. In our analysis we assume that the identification system is designed using a pilot-based re-synchronization mechanism that assists desynchronization compensation with a certain accuracy. We demonstrate how the accuracy of re-synchronization impacts the information-theoretic limits of identification system performance.

Content identification based on digital fingerprinting attracts a lot of attention in different emerging applications. In this paper, we consider digital identification based on the sign-magnitude decomposition of fingerprint codewords and analyze the achievable rates for each component. We introduce a channel splitting approach and reveal certain interesting phenomena related to channel polarization. It is demonstrated that under certain conditions almost all rate in the sign channel is concentrated in reliable components, this can be of interest for complexity and security in various content identification applications. The envisioned extensions cover applications where the input and output alphabets of the channel are different at the encoding and decoding stages. Additionally, the reduction of the input data dimensionality at the encoding/enrollment stage can increase the cryptographic protection in terms of privacy leakage and simplify the decoding algorithms in biometric applications.

In this work we advocate an approach for the statistical performance analysis of an identification system. The statistical performance analysis is accomplished for the corresponding probability of miss and false acceptance based on the order statistic list decoding framework.

In recent years, content identification based on digital fingerprinting attracts a lot of attention in different emerging applications. At the same time, the theoretical analysis of digital fingerprinting systems for finite length case remains an open issue. Additionally, privacy leaks caused by fingerprint storage, distribution and sharing in a public domain via third party outsourced services cause certain concerns in the cryptographic community. In this paper, we perform an information-theoretic analysis of finite length digital fingerprinting systems in a private content identification setup and reveal certain connections between fingerprint based content identification and Forney's erasure/list decoding. Along this analysis, we also consider complexity issues of fast content identification in large databases on remote untrusted servers.

In this paper, the performance of the content identification based on digital fingerprinting and order statistic list decoding is analyzed by evaluating the probabilities of correct identification, false acceptance and the probability mass function of queried binary fingerprint position on the list of candidates. The particular attention is dedicated to the cases when traditional maximum likelihood decoder fails to produce the reliable content identification. The maximum likelihood decoding is shown to be a particular case of order statistic list decoding for the list size equals 1. We demonstrate the efficiency of the proposed content identification system performance by investigating the probability mass function behavior and imposing the constraint on the cardinality of list size.

In light of the recent development of multimedia and networking technologies, an exponentially increasing amount of content is available via various public services. That is why content identification attracts a lot of attention. One possible technology for content identification is based on digital fingerprinting. When trying to establish information-theoretic limits in this application, usually it is assumed that the codewords are of infinite length and that a jointly typical decoder is used in the analysis. These assumptions represent a certain over-generalization for the majority of practical applications. Consequently, the impact of the finite length on the mentioned limits remains an open and largely unexplored problem. Furthermore, leaking of privacy-related information to third parties due to storage, distribution and sharing of fingerprinting data represents an emerging research issue that should be addressed carefully. This paper contains an information-theoretic analysis of finite length digital fingerprinting under privacy constraints. A particular link between the considered setup and Forney's erasure/list decoding [1] is presented. Finally, complexity issues of reliable identification in large databases are addressed.