Look Twice: Training-Free Evidence Highlighting in Multimodal Large Language Models

Answering questions about images often requires combining visual understanding with external knowledge. Multimodal Large Language Models (MLLMs) provide a natural framework for this setting, but they often struggle to identify the most relevant visual and textual evidence when answering knowledge-intensive queries. In such scenarios, models must integrate visual cues with retrieved textual evidence that is often noisy or only partially relevant, while also localizing fine-grained visual information in the image. In this work, we introduce Look Twice (LoT), a training-free inference-time framework that improves how pretrained MLLMs utilize multimodal evidence. Specifically, we exploit the model attention patterns to estimate which visual regions and retrieved textual elements are relevant to a query, and then generate the answer conditioned on this highlighted evidence. The selected cues are highlighted through lightweight prompt-level markers that encourage the model to re-attend to the relevant evidence during generation. Experiments across multiple knowledge-based VQA benchmarks show consistent improvements over zero-shot MLLMs. Additional evaluations on vision-centric and hallucination-oriented benchmarks further demonstrate that visual evidence highlighting alone improves model performance in settings without textual context, all without additional training or architectural modifications. Source code will be publicly released.

Decoding of Variable Length PLH Codes

In this paper, we address the problem of the decoding of variable length physical layer header (PLH) codes. We take as a case study the DVB-S2X standard, and, after selecting a suitable PLH code with variable codeword length, we propose two alternative noncoherent decoding strategies. The proposed decoding strategies allow to estimate the length of the transmitted PLH field, which is unknown at the receiver, jointly with the decoding of the PLH code. We demonstrate that it is possible to achieve an excellent decoding performance while significantly reducing the overhead due to the transmission of the PLH field with respect to a standard fixed-length PLH.