Conducting a SLR is especially useful when starting a new line of research, as it involves a detailed analysis of the research topic supported by the appropriate references. This type of secondary study should be conducted following a strict protocol to ensure quality and allow replication (Booth et al., 2016). Within the SLR process, manual and automated searches are performed to identify research papers related to the topic under review (Kitchenham and Charters, 2007). Therefore, the selection of primary studies, i.e., papers of sufficient quality and truly relevant to the topic, is one of the most important steps. It is also a time-consuming task due to potentially large search results if the queries are too open-ended or the research topic is too broad. Recently, artificial intelligence (AI) has emerged as a way to assist researchers in this task, as well as in other stages of the SLR process (de la Torre-López et al., 2023). The topic has gained even more relevance since the appearance of Large Language Models (LLMs) (Han et al., 2024; Galli et al., 2025). LLMs have expanded the capabilities of AI-assisted SLRs with the ability to extract information from papers, synthesise their findings and generate texts to accelerate SLR reporting.

Hoffman et al (2022)'s Chinchilla paper introduced the principle of compute-optimal scaling, laying a foundation for future scaling of language models. In the years since, however, valid concerns about Chinchilla have been raised: wide confidence intervals, discrepancies between its three approaches, and incongruities with other scaling laws. This raises a critical question for the field: Can practitioners still rely on Chinchilla's prescriptions? Our work demonstrates the answer is yes. We begin by uncovering that the model parameters central to Chinchilla's analyses were ambiguous: three interpretations are possible, with relative differences between different interpretations of model parameters as high as 15.2%. We find that, perhaps surprisingly, which model parameters are used for the analyses do not meaningfully affect key results: the scaling law estimates and the compute-optimal tokens-to-parameter ratio. Indeed, under one interpretation, the tokens-to-parameter ratio becomes more constant with the target compute budget. We then ask how distorted the Chinchilla model parameters could have been without meaningfully affecting the key results. By deliberately perturbing model parameters in four structured ways, we find that key Chinchilla results are most sensitive to additive or systematic errors, which can alter the otherwise flat trend of the optimal tokens-to-parameter ratio, but overall, Chinchilla's key results withstand sizable perturbations. Altogether, our findings offer the field renewed confidence in Chinchilla as a durable guide for scaling language models.

Reinforcement Learning with Verifiable Rewards (RLVR) for large language models (LLMs) has achieved remarkable progress in enhancing LLMs' reasoning capabilities on tasks with clear correctness criteria, such as mathematical reasoning tasks. Several training metrics, such as entropy or response length, have been observed to correlate with different reasoning behaviors in reinforcement learning. Prior approaches incorporate such priors through reward or advantage shaping, which often relies on hand-crafted penalties and preferences (e.g., higher-is-better or lower-is-better). However, without careful hyperparameter tuning, these directional priors can be overly biased and may lead to failure. To this end, we introduce Conditional advANtage estimatiON (CANON), amplifying the impact of the target metric without presuming its direction. Specifically, CANON regroups the sampled responses into two groups based on the higher or lower value of a target metric, measures which metric trend contributes to better performance through inter-group comparison, and identifies the better response within the same group. In summary, CANON based on entropy consistently outperforms prior methods across three LLMs on both math reasoning and high-complexity logic tasks. When applied to response length, CANON further improves token efficiency, yielding a more favorable Pareto frontier in the performance-cost trade-off.

Full-duplex interaction is crucial for natural human-machine communication, yet remains challenging as it requires robust turn-taking detection to decide when the system should speak, listen, or remain silent. Existing solutions either rely on dedicated turn-taking models, most of which are not open-sourced. The few available ones are limited by their large parameter size or by supporting only a single modality, such as acoustic or linguistic. Alternatively, some approaches finetune LLM backbones to enable full-duplex capability, but this requires large amounts of full-duplex data, which remain scarce in open-source form. To address these issues, we propose Easy Turn, an open-source, modular turn-taking detection model that integrates acoustic and linguistic bimodal information to predict four dialogue turn states: complete, incomplete, backchannel, and wait, accompanied by the release of Easy Turn trainset, a 1,145-hour speech dataset designed for training turn-taking detection models. Compared to existing open-source models like TEN Turn Detection and Smart Turn V2, our model achieves state-of-the-art turn-taking detection accuracy on our open-source Easy Turn testset. The data and model will be made publicly available on GitHub.

Prior work has largely treated future event prediction as a static task, failing to consider how forecasts and the confidence in them should evolve as new evidence emerges. To address this gap, we introduce EVOLVECAST, a framework for evaluating whether large language models appropriately revise their predictions in response to new information. In particular, EVOLVECAST assesses whether LLMs adjust their forecasts when presented with information released after their training cutoff. We use human forecasters as a comparative reference to analyze prediction shifts and confidence calibration under updated contexts. While LLMs demonstrate some responsiveness to new information, their updates are often inconsistent or overly conservative. We further find that neither verbalized nor logits-based confidence estimates consistently outperform the other, and both remain far from the human reference standard. Across settings, models tend to express conservative bias, underscoring the need for more robust approaches to belief updating.

Mixture-of-Experts (MoE) architectures have emerged as a promising direction, offering efficiency and scalability by activating only a subset of parameters during inference. However, current research remains largely performance-centric, with limited understanding of its internal mechanisms, thereby constraining broader progress. In this work, we use an internal metric to investigate the mechanisms of MoE architecture by explicitly incorporating routing mechanisms and analyzing expert-level behaviors. Through systematic analyses of a wide range of publicly available MoE models, we uncover several findings: (1) neuron utilization decreases as models evolve, reflecting stronger generalization; (2) training exhibits a dynamic trajectory, where benchmark performance alone provides limited signal while MUI reveals deeper insights; (3) task completion emerges from collaborative contributions of multiple experts, with shared experts driving concentration; and (4) activation patterns at the neuron level provide a fine-grained proxy for data diversity. Together, these results demonstrate the potential of MUI as a complementary indicator to benchmark performance, offering new insights into the capacity, dynamics, and specialization of MoE models. Our project can be found at https://yingjiahao14.github.io/MoE-MUI/.

Masked diffusion language models (MDLMs) have recently emerged as a promising alternative to autoregressive (AR) language models, offering properties such as parallel decoding, flexible generation orders, and the potential for fewer inference steps. Despite these advantages, decoding strategies and reinforcement learning (RL) algorithms tailored for MDLMs remain underexplored. A naive approach is to directly transfer techniques well-established for AR models to MDLMs. However, this raises an immediate question: Is such a naive transfer truly optimal? For example, 1) Block-wise and semi-AR decoding strategies are not employed during the training of MDLMs, so why do they outperform full diffusion-style decoding during inference? 2) Applying RL algorithms designed for AR models directly to MDLMs exhibits a training-inference inconsistency, since MDLM decoding are non-causal (parallel). This results in inconsistencies between the rollout trajectory and the optimization trajectory. To address these challenges, we propose EOS Early Rejection (EOSER) and Ascending Step-Size (ASS) decoding scheduler, which unlock the potential of MDLMs to perform full diffusion-style decoding, achieving competitive performance with fewer decoding steps. Additionally, we introduce Consistency Trajectory Group Relative Policy Optimization (CJ-GRPO) for taming MDLMs, which emphasizes the consistency between rollout trajectory and optimization trajectory, and reduces the optimization errors caused by skip-step optimization. We conduct extensive experiments on reasoning tasks, such as mathematical and planning benchmarks, using LLaDA-8B-Instruct. The results demonstrate that the proposed EOSER and ASS mechanisms, together with CJ-GRPO, hold significant promise for effectively and efficiently taming MDLMs. Code: https://github.com/yjyddq/EOSER-ASS-RL.

Catastrophic forgetting remains a critical challenge in continual learning for large language models (LLMs), where models struggle to retain performance on historical tasks when fine-tuning on new sequential data without access to past datasets. In this paper, we first reveal that the drift of functional directions during the fine-tuning process is a key reason why existing regularization-based methods fail in long-term LLM continual learning. To address this, we propose Dynamic Orthogonal Continual (DOC) fine-tuning, a novel approach that tracks the drift of these functional directions and dynamically updates them during the fine-tuning process. Furthermore, by adjusting the gradients of new task parameters to be orthogonal to the tracked historical function directions, our method mitigates interference between new and old tasks. Extensive experiments on various LLM continual learning benchmarks demonstrate that this approach outperforms prior methods, effectively reducing catastrophic forgetting and providing a robust tool for continuous LLM fine-tuning. Our code is available at https://github.com/meloxxxxxx/DOC.

Language models can transfer hidden biases during distillation. For example, a teacher that "likes owls" can make its student "like owls" too, even when the training data consists only of lists of numbers. This surprising phenomenon is called subliminal learning. Subliminal learning can be expected under soft distillation, where the student is trained on the teacher's full next-token distribution. But the fact that this also occurs under hard distillation--where the student only sees sampled tokens--raises a deeper question: when and how does subliminal learning actually occur? We answer this question through controlled experiments and mechanistic analysis. Our results show that subliminal learning does not need (global) token entanglement or logit leakage. Instead, it comes down to a small set of divergence tokens--rare cases where teachers with different biases would predict different tokens. Masking out these tokens mostly removes the hidden bias transfer. Mechanistically, divergence tokens reveal that early layers are critical. Surprisingly, finetuning even a single such early layer is sufficient for subliminal learning. Finally, we find that subliminal learning is fragile. Even small changes, like paraphrasing prompts, are usually sufficient to suppress it. Distillation is a core technique for compressing models or transferring knowledge, where a student model is trained to imitate a teacher (Hinton et al., 2015; Ba & Caruana, 2014). The common view is that what transfers depends on the (semantic) content of the training data (Dong et al., 2023; Guan et al., 2024; Chen et al., 2025; Li et al., 2025). In this view, if the teacher's outputs do not show a trait--such as a bias toward an animal or misaligned behavior--the student should not learn it.

The reliability of Multimodal Large Language Models (MLLMs) in real-world settings is often undermined by sensitivity to irrelevant or distracting visual context, an aspect not captured by existing evaluation metrics. We introduce the \textbf{Patch Context Robustness Index (PCRI)}, the first systematic and interpretable score for quantifying MLLM robustness to variations in visual context granularity, measuring performance changes between localized image patches and full-image input. Applying PCRI to 19 state-of-the-art MLLMs across 15 vision-language benchmarks, we find that most leading models remain brittle to background noise, with only a few, such as InternVL2-26B and Qwen2VL-72B, demonstrating consistent robustness across tasks. PCRI analysis also highlights how different model architectures handle and integrate visual context, offering actionable diagnostic insight for both researchers and practitioners. PCRI enables rigorous comparison of context robustness, supporting principled model selection and guiding the development of future architectures and training strategies for robust, real-world deployment.