The experiments, reported in a 2026 research article, used CIFAR-10, CIFAR-100, DIV2K and Kodak24 images. CIFAR-10 supplied 50,000 training and 10,000 test images at 32 by 32 pixels. DIV2K supplied 800 training and 100 test images cropped to 256 by 256 pixels, while the 24 Kodak24 images, measuring 768 by 512 pixels, were used only for evaluation. These were simulation tests rather than trials on a physical wireless network or real-world radio traces.

CLEAR contains a semantic encoder, a simulated channel module, the adaptive diffusion denoising model and a semantic decoder. The encoder converts an image into a compact representation that passes through the channel. At the receiver, the denoising model iteratively estimates and removes noise before the decoder reconstructs the image. The channel model included six multipath components, Doppler shifts, phase noise and additive Gaussian noise, with a carrier frequency of 2.4 GHz.

Unlike methods that adapt mainly to signal-to-noise ratio, CLEAR uses a structured channel embedding derived from estimated impulse-response statistics, Doppler shift and phase-noise level. A lightweight multilayer perceptron converts this information into an embedding, which modulates the denoising network at every diffusion step through feature-wise linear modulation. The reported results used estimated rather than perfect channel state information: a zero-mean Gaussian error with a standard deviation equal to 5% of each parameter’s dynamic range was added to the channel information used during training and evaluation.

All models were trained for 200 epochs with a batch size of 128 and Adam optimization. The learning rate began at 10−4 and was halved every 50 epochs. The diffusion schedule used 50 steps in both training and inference, while training signal-to-noise ratios ranged from 1 to 25 dB and testing covered 0 to 30 dB. Compression ratios of 0.3, 0.5 and 0.6 were evaluated. Experiments were repeated five times with different random seeds, and results were reported as means with standard deviations; the stated statistical comparison used a paired t-test.

Under complex time-varying conditions with a Doppler parameter of 0.5, phase-noise parameter of 0.5 and compression ratio of 0.5, CLEAR achieved the best reported results on all three metrics. On DIV2K, it improved on DeepJSCC-V by 2.31 dB in peak signal-to-noise ratio, 0.041 in structural similarity and 0.041 in learned perceptual image patch similarity. The PSNR difference was reported as statistically significant at p<0.01. CLEAR also led SCAN by 2.05 dB in PSNR on that dataset.

The pattern extended to other image sets. On CIFAR-100, CLEAR recorded 24.12 dB PSNR, 2.19 dB above DeepJSCC-V and 1.94 dB above SCAN. On Kodak24, it reached 27.13 dB PSNR, a 2.32 dB improvement over DeepJSCC-V. Reconstruction quality generally increased with signal-to-noise ratio and fell as Doppler shift and phase-noise levels increased. The authors also observed a temporary performance dip near a test signal-to-noise ratio of 5 dB in some curves and attributed it to mismatch between training and test channel statistics, although they described that mechanism as an interpretation requiring further study.

Ablation results indicated that the diffusion denoiser was more effective than a five-layer residual CNN denoiser of comparable parameter count. With 50 diffusion steps, the adaptive model exceeded the CNN alternative by 1.56 dB PSNR and 0.027 SSIM while using fewer parameters. A 10-step version retained a 0.81 dB advantage and added 1.7 ms of latency. Removing individual channel-information components reduced performance: the reported losses relative to full conditioning were 0.63 dB without Doppler shift, 0.51 dB without phase noise and 0.43 dB without impulse-response statistics.

The principal limitation is the reliance on simulated channel distributions. The study did not evaluate CLEAR on a deployed radio system or real-world channel traces, and its conclusions therefore apply to the tested datasets, compression ratios and modeled impairments. Inference also has a cost: the 50-step version required 12.8 milliseconds on an NVIDIA RTX 3090 for 256-by-256-pixel images, approximately 5.6 times the DeepJSCC-V latency, although this measurement excluded data loading, preprocessing and channel-state estimation. The authors identify testing with software-defined-radio traces, extending the method to audio and video, and reducing latency as future work.

Subject of Research: Simulation testing of a channel-aware semantic communication framework for wireless image reconstruction

Article Title: Channel learning and enhanced adaptive reconstruction for robust image transmission over complex time varying wireless channels

Article References: Pan, H., Zu, T., Wang, L., Wu, S., & He, X. (2026). Channel learning and enhanced adaptive reconstruction for robust image transmission over complex time varying wireless channels. Discover Artificial Intelligence, 6(1), Article 1453. https://doi.org/10.1007/s44163-026-02499-z

Image Credits: AI Generated

DOI: 10.1007/s44163-026-02499-z

Keywords: wireless image transmission, semantic communication, diffusion denoising, deep joint source–channel coding, channel state information, Doppler shift, phase noise, image reconstruction, multipath fading, simulation study, Channel, learning

The framework, called MVHD for Multi-View Hypergraph Disentanglement and diffusion denoising, was developed by Huan Li, Xianying Huang and Xiaoyun Ma at the College of Computer Science and Engineering, Chongqing University of Technology in Chongqing, China. In their paper, the authors identify two core challenges that plague existing conversational recommenders. The first is the naive fusion of multi-source heterogeneous information. Modern systems draw preference signals from many places simultaneously: the user’s interaction history with items, knowledge graphs that encode attributes and relationships between entities, and the linguistic content of the conversation itself. When these disparate signals are simply concatenated or averaged into a single user vector, redundant and contradictory information creeps in. One source may suggest a user prefers action films, another may hint at a taste for documentaries, and a third may reflect only an artifact of how the data was collected. The resulting representation is a blurred composite that limits fine-grained characterization of what the user actually wants.

The second challenge concerns robustness. Real conversations are messy. Users speak casually, use shorthand, change their minds mid-sentence, and refer to things obliquely. Knowledge graphs, meanwhile, are far from clean; they contain incomplete links, erroneous triples, and noise inherited from automated extraction pipelines. Colloquial expressions and graph noise alike corrode the quality of user representations, and because both the recommendation engine and the dialogue generator consume that same representation, the damage propagates to every part of the system’s behavior. A slightly corrupted preference vector can lead the system to recommend an irrelevant item and then to justify that recommendation with an incoherent or factually wrong sentence.

To address these problems, MVHD begins by constructing three complementary hypergraph views of user preferences. Hypergraphs are a generalization of ordinary graphs in which a single edge, called a hyperedge, can connect any number of nodes at once. This property makes them naturally suited to modeling group phenomena in recommendation data: a set of users who all interacted with the same cluster of items, or a group of entities sharing a common attribute, can be captured by one hyperedge rather than by a tangled web of pairwise links. The first view built by the authors is the collaborative attribute view, which links users through the attributes of the items they have engaged with. The second is the structural semantic view, which encodes relationships drawn from the knowledge graph, capturing how entities and concepts connect semantically. The third is the similar user groups view, which captures communities of users whose behavioral patterns resemble one another. Each view, in isolation, offers a distinct and structured lens on preference; together they span the behavioral, semantic and social dimensions of taste.

But simply having three views does not solve the fusion problem; it can make it worse. The crucial innovation lies in what the authors do next: a dynamic feature disentanglement mechanism. Rather than mixing the three views wholesale, the system explicitly separates, within each view, the features that are common across views from the features that are specific to that view. Common features represent stable, consistent signals of preference that recur regardless of which source produced them, while view-specific features capture the unique contribution of each perspective. By extracting these two categories separately and then performing a personalized integration tailored to each individual user, the framework assembles an initial user representation that is well disentangled from the start. In effect, the system learns which parts of a user’s profile are corroborated by multiple evidence streams and which parts are idiosyncratic to a single stream, and it keeps them distinguishable rather than allowing them to blur together.

The second half of the framework tackles noise with a tool borrowed from generative modeling: the diffusion process. Diffusion models, which have revolutionized image generation, work by gradually adding noise to data and then learning to reverse that process step by step. MVHD adapts this idea for representation purification through a condition-guided diffusion denoising strategy. In the forward process, controlled noise is deliberately injected into the initial disentangled user representation. In the reverse process, a denoising network learns to strip that noise away, but it does not do so blindly. The reverse denoising is conditioned on collaborative attribute information, which serves as a semantic guide, steering the reconstruction toward representations that are consistent with the user’s known attribute profile. The final output is a user representation that has been passed through this noise-injection and purification cycle and emerges more robust and refined than the version that entered it.

The choice of collaborative attributes as the conditioning signal is deliberate. Attributes attached to items and entities tend to be more stable than interaction counts or conversational cues, which fluctuate with session context and phrasing. By anchoring the denoising process to this stable substrate, the model gains a reference point that survives the noise present in the other sources. The authors describe this as an explicit purification of the initial representation, in contrast to approaches that hope robustness emerges implicitly from training on large datasets.

The framework was evaluated extensively on two benchmark datasets commonly used in conversational recommendation research, covering both of the system’s dual tasks. On the recommendation side, where the goal is to surface the correct entity or item from the conversation, MVHD significantly outperformed existing state-of-the-art methods. On the conversation side, where the system must generate natural-language responses, it likewise surpassed competing approaches, indicating that a cleaner user representation benefits not only the ranking of candidate items but also the quality and informational density of the dialogue itself.

A case study included in the paper’s appendix illustrates the difference concretely. When compared with representative baselines such as UniCRS and MPKE, the responses generated by MVHD were judged more fluent and more accurate. The authors attribute this to the richer knowledge information flowing through the model, but they also note a specific advantage: because MVHD incorporates collaborative attribute information, it can weave more entity-related attributes into its generated sentences than MPKE can. The result is generated content that is more comprehensive and information-dense, integrating multi-source attribute information rather than relying on a single semantic channel.

The study also offers transparency into its use of large language models. To extract what the authors call Preference Evolution Information from dialogues, the researchers employed a 4-bit quantized Gemma3 12-billion-parameter model operating under a structured prompt template, with decoding hyperparameters set to a temperature of 0.8, top-p of 0.9 and top-k of 30. They acknowledge that autoregressive language models, particularly when quantized, introduce some run-to-run variability in the exact wording of extracted preferences, but their empirical observations indicate that the core semantic structures of the preference shifts remain highly stable and consistently effective for the downstream multi-view hypergraph encoding. This detail matters for reproducibility, and the authors state that the datasets, code and prompts used in the study are available from the corresponding author upon reasonable request.

The broader significance of the work lies in its treatment of representation quality as the shared foundation of recommendation and conversation. Much of the recent literature on conversational recommenders has focused on architectural refinements to the dialogue component, prompted in part by the arrival of powerful large language models that can produce impressively fluent text. But fluency without grounded, accurate preference understanding can produce confident nonsense, and noisy or entangled user vectors undermine even the most capable language backbone. By combining hypergraph-based multi-view structural modeling, explicit disentanglement of common versus view-specific features, and diffusion-based denoising guided by collaborative attributes, MVHD offers a systematic pipeline for building user representations that are simultaneously expressive, modular and noise-resistant.

The research was supported in part by the National Natural Science Foundation of China under Grant No. 62141201 and by the Chongqing Banan District Science and Technology Bureau under Grant No. 2020QC403. The paper was received in February 2026, revised in June, and published on 23 June 2026. As conversational interfaces increasingly replace search boxes in e-commerce, media and information services, frameworks like MVHD point toward systems that not only talk the talk but genuinely understand, with structural precision, who they are talking to.