Searcharxiv⌕ Search

arXiv subjects

Jithin Pradeep

Publications and source records attributed to Jithin Pradeep.

6 recordsLinked to original sources

Right Token, Right Time: From Per-Token Gating to Trajectory-Aware Activation Steering

Activation steering via sparse autoencoders (SAEs) enables behavioral control of large language models without task-specific fine-tuning. Yet, standard methods apply the steering signal at every generated token, displacing representations from the distribution the model was trained on and eroding generation quality. We show that \emph{when} along a generation the intervention is applied is a decisive axis of control. We cast steering as a per-token allocation problem under a fixed budget of steerable tokens, so that methods differ only in which tokens they select, making allocation the sole variable. Within this framework we introduce two gates. The first, Contrastive Feature Projection (CFP), is memoryless and selects the tokens least aligned with the target behavior through a sparse, interpretable feature projection. The second, Closed-Loop Adaptive Bernoulli Steering (CLABS), is trajectory aware. It treats allocation as closed-loop feedback control, tracking the alignment a generation has accumulated so far and spending budget where the trajectory falls behind. On a linear steering-response model we derive an explicit gain region that guarantees closed-loop stability of CLABS and show that its integral term reduces the steady-state tracking offset. Empirically, CFP and CLABS achieve the largest behavioral shift among budgeted gates at matched budgets, with CLABS strongest under the tight budgets where allocation matters most, while best preserving fluency.

cs.LG↗

Reject, Remember, Retry: Why Multi-Turn Attackers Should Remember Their Mistakes

Existing multi-turn jailbreak attackers condition on the same conversation history as the target. We introduce $R^3$ (Reject, Remember, Retry), a black-box multi-LLM framework in which the attacker conditions on a compact state rather than the raw transcript. Turns that fail to raise the attack risk score are removed from the target's context but remembered by the attacker, together with the reason for failure. We call this asymmetric memory. Across two benchmarks and six target LLMs, $R^3$ outperforms state-of-the-art baselines by up to 53% in ASR@5 under matched query budgets, and a low-capacity attacker with $R^3$'s state is able to outperform high-capacity attackers without it by up to 34% in ASR@5. Ablations attribute most gains to remembering rejected turns, and the risk score acts as a coarse threshold rather than a progress measure.

cs.AI↗

The Format is the Filter: Scoring Format Fixes What an LLM Judge Can Express

When an LLM judge filters instruction-tuning data, its scoring format (binary verdict, 1--5 rating, 0--100 score, or comparison) is rarely justified or ablated. Holding the candidate pool, judge prompt, decoding and selection budget fixed, we score 15,406 UltraFeedback candidates under five formats with three judges, curate, fine-tune 66 models, and evaluate them with four judges. A 0--100 scale is worth only 7--12 equally used levels, two to three times a 1--5 scale rather than twenty, and a binary verdict leaves the top score tied in 77--78\% of prompts, so most selections are made at random. These properties are set by the format almost independently of the judge. Kendall's $τ_b$ hides them: it ranks binary best while $τ_a$ and top-1 agreement rank it last, because it discounts the ties a selector must break at random. Both comparative formats inherit a slot bias whose sign differs across judges. Downstream, curated data beat random selection under every evaluation judge, robust to length and markdown adjustment, but no ordering among formats is detectable. We propose four quantities that judge-based curation pipelines should report.

cs.AI↗

A Coin Flip Per Token: Bernoulli Sparse Steering of Large Language Models

Activation steering via sparse autoencoders (SAEs) enables behavioral control of large language models without task-specific fine-tuning, but standard methods apply the steering signal at every generated token, incurring constant per-token perturbation that risks degrading fluency. We ask: is dense intervention necessary? We introduce Stochastic Token Steering (STS), which gates each token independently with probability $p$, and Stochastic Block Steering (SBS), which gates a leading window once per sequence; neither requires a reward model or learned gating policy. Across two model families and two behavioral tasks, steering only a fraction of tokens recovers most of the behavioral effect of dense steering while incurring less fluency degradation. Across our experimental settings, we further find that the best-performing steering magnitude consistently rises as the intervention ratio falls.

cs.LG↗

Learning Robust Latent Representations for Controllable Speech Synthesis

State-of-the-art Variational Auto-Encoders (VAEs) for learning disentangled latent representations give impressive results in discovering features like pitch, pause duration, and accent in speech data, leading to highly controllable text-to-speech (TTS) synthesis. However, these LSTM-based VAEs fail to learn latent clusters of speaker attributes when trained on either limited or noisy datasets. Further, different latent variables start encoding the same features, limiting the control and expressiveness during speech synthesis. To resolve these issues, we propose RTI-VAE (Reordered Transformer with Information reduction VAE) where we minimize the mutual information between different latent variables and devise a modified Transformer architecture with layer reordering to learn controllable latent representations in speech data. We show that RTI-VAE reduces the cluster overlap of speaker attributes by at least 30\% over LSTM-VAE and by at least 7\% over vanilla Transformer-VAE.

cs.SD↗

Machine Learning Fund Categorizations

Given the surge in popularity of mutual funds (including exchange-traded funds (ETFs)) as a diversified financial investment, a vast variety of mutual funds from various investment management firms and diversification strategies have become available in the market. Identifying similar mutual funds among such a wide landscape of mutual funds has become more important than ever because of many applications ranging from sales and marketing to portfolio replication, portfolio diversification and tax loss harvesting. The current best method is data-vendor provided categorization which usually relies on curation by human experts with the help of available data. In this work, we establish that an industry wide well-regarded categorization system is learnable using machine learning and largely reproducible, and in turn constructing a truly data-driven categorization. We discuss the intellectual challenges in learning this man-made system, our results and their implications.

q-fin.ST↗