Large language models (LLMs) increasingly solve complex reasoning tasks via long chain-of-thought, but their forward-only autoregressive generation process is fragile; early token errors can cascade, which creates a clear need for self-reflection mechanisms. However, existing self-reflection either performs revisions over full drafts or learns self-correction via expensive training, both fundamentally reactive and inefficient. To address this, we propose Self-Reflective Generation at Test Time (SRGen), a lightweight test-time framework that reflects before generating at uncertain points. During token generation, SRGen utilizes dynamic entropy thresholding to identify high-uncertainty tokens. For each identified token, it trains a specific corrective vector, which fully exploits the already generated context for a self-reflective generation to correct the token probability distribution. By retrospectively analyzing the partial output, this self-reflection enables more trustworthy decisions, thereby significantly reducing the probability of errors at highly uncertain points. Evaluated on challenging mathematical reasoning benchmarks and a diverse set of LLMs, SRGen can consistently strengthen model reasoning: improvements in single-pass quality also translate into stronger self-consistency voting. The ability to execute complex multi-step reasoning remains a central frontier in advancing large language models (LLMs). LLMs generate step-by-step reasoning traces, often called chain-of-thought (CoT) (Wei et al., 2022). This capability has enabled substantial progress in mathematics, program synthesis, and other domains (Y ao et al., 2023; Plaat et al., 2024). The fidelity of these traces often determines whether the final answer is correct (Paul et al., 2024; Hammoud et al., 2025). Thus, improving the reliability of the reasoning process is critical to realizing the full potential of LLMs.

As Large Language Models become integral to software development, with substantial portions of AI-suggested code entering production, understanding their internal correctness mechanisms becomes critical for safe deployment. We apply sparse autoencoders to decompose LLM representations, identifying directions that correspond to code correctness. We select predictor directions using t-statistics and steering directions through separation scores from base model representations, then analyze their mechanistic properties through steering, attention analysis, and weight orthogonalization. We find that code correctness directions in LLMs reliably predict incorrect code, while correction capabilities, though statistically significant, involve tradeoffs between fixing errors and preserving correct code. Mechanistically, successful code generation depends on attending to test cases rather than problem descriptions. Moreover, directions identified in base models retain their effectiveness after instruction-tuning, suggesting code correctness mechanisms learned during pre-training are repurposed during fine-tuning. Our mechanistic insights suggest three practical applications: prompting strategies should prioritize test examples over elaborate problem descriptions, predictor directions can serve as error alarms for developer review, and these same predictors can guide selective steering, intervening only when errors are anticipated to prevent the code corruption from constant steering.

Multi-hop reasoning over financial disclosures is often a retrieval problem before it becomes a reasoning or generation problem: relevant facts are dispersed across sections, filings, companies, and years, and LLMs often expend excessive tokens navigating noisy context. Without precise Knowledge Graph (KG)-guided selection of relevant context, even strong reasoning models either fail to answer or consume excessive tokens, whereas KG-linked evidence enables models to focus their reasoning on composing already retrieved facts. We present FinReflectKG - MultiHop, a benchmark built on FinReflectKG, a temporally indexed financial KG that links audited triples to source chunks from S&P 100 filings (2022-2024). Mining frequent 2-3 hop subgraph patterns across sectors (via GICS taxonomy), we generate financial analyst style questions with exact supporting evidence from the KG. A two-phase pipeline first creates QA pairs via pattern-specific prompts, followed by a multi-criteria quality control evaluation to ensure QA validity. We then evaluate three controlled retrieval scenarios: (S1) precise KG-linked paths; (S2) text-only page windows centered on relevant text spans; and (S3) relevant page windows with randomizations and distractors. Across both reasoning and non-reasoning models, KG-guided precise retrieval yields substantial gains on the FinReflectKG - MultiHop QA benchmark dataset, boosting correctness scores by approximately 24 percent while reducing token utilization by approximately 84.5 percent compared to the page window setting, which reflects the traditional vector retrieval paradigm. Spanning intra-document, inter-year, and cross-company scopes, our work underscores the pivotal role of knowledge graphs in efficiently connecting evidence for multi-hop financial QA. We also release a curated subset of the benchmark (555 QA Pairs) to catalyze further research.

Reinforcement learning (RL) is central to improving reasoning in large language models (LLMs) but typically requires ground-truth rewards. Test-Time Reinforcement Learning (TTRL) removes this need by using majority-vote rewards, but relies on heavy online RL and incurs substantial computational cost. We propose RoiRL: Reasoning with offline iterative Reinforcement Learning, a family of lightweight offline learning alternatives that can target the same regularized optimal policies. Unlike TTRL, RoiRL eliminates the need to maintain a reference model and instead optimizes weighted log-likelihood objectives, enabling stable training with significantly lower memory and compute requirements. Experimental results show that RoiRL trains to 2.5x faster and consistently outperforms TTRL on reasoning benchmarks, establishing a scalable path to self-improving LLMs without labels.

Zero-shot Text-to-Speech (TTS) has recently advanced significantly, enabling models to synthesize speech from text using short, limited-context prompts. These prompts serve as voice exemplars, allowing the model to mimic speaker identity, prosody, and other traits without extensive speaker-specific data. Although recent approaches incorporating language models, diffusion, and flow matching have proven their effectiveness in zero-shot TTS, they still encounter challenges such as unreliable synthesis caused by token repetition or unexpected content transfer, along with slow inference and substantial computational overhead. Moreover, temporal diversity-crucial for enhancing the naturalness of synthesized speech-remains largely underexplored. To address these challenges, we propose Flamed-TTS, a novel zero-shot TTS framework that emphasizes low computational cost, low latency, and high speech fidelity alongside rich temporal diversity. To achieve this, we reformulate the flow matching training paradigm and incorporate both discrete and continuous representations corresponding to different attributes of speech. Experimental results demonstrate that Flamed-TTS surpasses state-of-the-art models in terms of intelligibility, naturalness, speaker similarity, acoustic characteristics preservation, and dynamic pace. Notably, Flamed-TTS achieves the best WER of 4% compared to the leading zero-shot TTS baselines, while maintaining low latency in inference and high fidelity in generated speech. Code and audio samples are available at our demo page https://flamed-tts.github.io.

A common but rarely examined assumption in machine learning is that training yields models that actually satisfy their specified objective function. We call this the Objective Satisfaction Assumption (OSA). Although deviations from OSA are acknowledged, their implications are overlooked. We argue, in a learning-paradigm-agnostic framework, that OSA fails in realistic conditions: approximation, estimation, and optimization errors guarantee systematic deviations from the intended objective, regardless of the quality of its specification. Beyond these technical limitations, perfectly capturing and translating the developer's intent, such as alignment with human preferences, into a formal objective is practically impossible, making misspecification inevitable. Building on recent mathematical results, absent a mathematical characterization of these gaps, they are indistinguishable from those that collapse into Goodhart's law failure modes under strong optimization pressure. Because the Goodhart breaking point cannot be located ex ante, a principled limit on the optimization of General-Purpose AI systems is necessary. Absent such a limit, continued optimization is liable to push systems into predictable and irreversible loss of control.

Although recent tool-augmented benchmarks incorporate complex user requests and diverse tools, the evaluation methods for most of them remain limited to answer matching. However, as the number of steps required to resolve a user request increases, a proper evaluation of an agent's performance must go beyond the final answer to also assess the problem-solving trajectory, including previously ignored aspects such as efficiency, hallucination, and adaptivity. The most straightforward method for evaluating these aspects is to compare an agent's trajectory with the ground-truth trajectory, but this approach is fundamentally limited since annotating all valid ground-truth trajectories is prohibitively expensive. However, a simple LLM-based evaluator struggles to assess trajectories in detail without ground truth. To effectively evaluate the agents in this manner, we introduce TRACE, a framework for the multi-dimensional evaluation of tool-augmented LLM agent performance. By incorporating an evidence bank, which accumulates knowledge gathered from preceding reasoning steps, TRACE enables a multi-faceted analysis and evaluation of an agent's reasoning trajectory effectively. To validate our framework, we develop a new meta-evaluation dataset by augmenting existing benchmarks with diverse and flawed trajectories, each labeled with multi-faceted performance scores. Our results confirm that TRACE accurately evaluates these complex behaviors in a scalable and cost-effective manner, even with small open-source LLMs. Furthermore, we apply our method to evaluate the trajectories that agents produce while solving tool-augmented tasks, presenting previously unreported observations and their corresponding insights. As this field rapidly evolves, various benchmark datasets have been introduced to measure agent performance. However, while extensive evaluation has focused on what tasks an LLM agent can accomplish, the assessment of how it accomplishes them has been largely overlooked. Existing benchmarks predominantly rely on an Answer Match evaluation, which only verifies if the final output matches the ground-truth answer (Huang et al., 2023; Wang et al., 2024; Ma et al., 2024; Mialon et al., 2023; Zhang et al., 2024b).

Large Language Models (LLMs) are rapidly being adopted in conservation to address the biodiversity crisis, yet their reliability for species evaluation is uncertain. This study systematically validates five leading models on 21,955 species across four core IUCN Red List assessment components: taxonomy, conservation status, distribution, and threats. A critical paradox was revealed: models excelled at taxonomic classification (94.9%) but consistently failed at conservation reasoning (27.2% for status assessment). This knowledge-reasoning gap, evident across all models, suggests inherent architectural constraints, not just data limitations. Furthermore, models exhibited systematic biases favoring charismatic vertebrates, potentially amplifying existing conservation inequities. These findings delineate clear boundaries for responsible LLM deployment: they are powerful tools for information retrieval but require human oversight for judgment-based decisions. A hybrid approach is recommended, where LLMs augment expert capacity while human experts retain sole authority over risk assessment and policy.

Verifying multi-step reasoning in large language models is difficult due to imprecise error localization and high token costs. Existing methods either assess entire reasoning chains, suffering attention dilution, or rely on expensive multi-sampling. We introduce Node-wise Consistency Verification (NCV), a training-free framework that recasts verification as lightweight binary consistency checks at the node level. By decomposing the chain of thought into interconnected verification nodes, NCV precisely localizes errors and avoids unnecessary long-form generation. Experiments demonstrate that our approach enhances interpretability and efficiency, presenting a scalable solution for reliable LLM reasoning verification. On public datasets, NCV achieves a 10\% to 25\% improvement in F1 scores over baselines while utilizing $6\times$~$58\times$ fewer tokens than traditional methods like CoT-based verifiers.

Personality refers to individual differences in behavior, thinking, and feeling. With the growing availability of digital footprints, especially from social media, automated methods for personality assessment have become increasingly important. Natural language processing (NLP) enables the analysis of unstructured text data to identify personality indicators. However, two main challenges remain central to this thesis: the scarcity of large, personality-labeled datasets and the disconnect between personality psychology and NLP, which restricts model validity and interpretability. To address these challenges, this thesis presents two datasets -- MBTI9k and PANDORA -- collected from Reddit, a platform known for user anonymity and diverse discussions. The PANDORA dataset contains 17 million comments from over 10,000 users and integrates the MBTI and Big Five personality models with demographic information, overcoming limitations in data size, quality, and label coverage. Experiments on these datasets show that demographic variables influence model validity. In response, the SIMPA (Statement-to-Item Matching Personality Assessment) framework was developed - a computational framework for interpretable personality assessment that matches user-generated statements with validated questionnaire items. By using machine learning and semantic similarity, SIMPA delivers personality assessments comparable to human evaluations while maintaining high interpretability and efficiency. Although focused on personality assessment, SIMPA's versatility extends beyond this domain. Its model-agnostic design, layered cue detection, and scalability make it suitable for various research and practical applications involving complex label taxonomies and variable cue associations with target concepts.