Monica Sunkara

Saket Dingliwal, Monica Sunkara, Sravan Bodapati et al.

End-to-end speech recognition models trained using joint Connectionist Temporal Classification (CTC)-Attention loss have gained popularity recently. In these models, a non-autoregressive CTC decoder is often used at inference time due to its speed and simplicity. However, such models are hard to personalize because of their conditional independence assumption that prevents output tokens from previous time steps to influence future predictions. To tackle this, we propose a novel two-way approach that first biases the encoder with attention over a predefined list of rare long-tail and out-of-vocabulary (OOV) words and then uses dynamic boosting and phone alignment network during decoding to further bias the subword predictions. We evaluate our approach on open-source VoxPopuli and in-house medical datasets to showcase a 60% improvement in F1 score on domain-specific rare words over a strong CTC baseline.

Rishabh Bhardwaj, George Polovets, Monica Sunkara

Semi-parametric Nearest Neighbor Language Models ($k$NN-LMs) have produced impressive gains over purely parametric LMs, by leveraging large-scale neighborhood retrieval over external memory datastores. However, there has been little investigation into adapting such models for new domains. This work attempts to fill that gap and suggests the following approaches for adapting $k$NN-LMs -- 1) adapting the underlying LM (using Adapters), 2) expanding neighborhood retrieval over an additional adaptation datastore, and 3) adapting the weights (scores) of retrieved neighbors using a learned Rescorer module. We study each adaptation strategy separately, as well as the combined performance improvement through ablation experiments and an extensive set of evaluations run over seven adaptation domains. Our combined adaptation approach consistently outperforms purely parametric adaptation and zero-shot ($k$NN-LM) baselines that construct datastores from the adaptation data. On average, we see perplexity improvements of 17.1% and 16% for these respective baselines, across domains.

Nilaksh Das, Saket Dingliwal, Srikanth Ronanki et al. · amazon-science

Large language models (LLMs) have shown incredible proficiency in performing tasks that require semantic understanding of natural language instructions. Recently, many works have further expanded this capability to perceive multimodal audio and text inputs, but their capabilities are often limited to specific fine-tuned tasks such as automatic speech recognition and translation. We therefore develop SpeechVerse, a robust multi-task training and curriculum learning framework that combines pre-trained speech and text foundation models via a small set of learnable parameters, while keeping the pre-trained models frozen during training. The models are instruction finetuned using continuous latent representations extracted from the speech foundation model to achieve optimal zero-shot performance on a diverse range of speech processing tasks using natural language instructions. We perform extensive benchmarking that includes comparing our model performance against traditional baselines across several datasets and tasks. Furthermore, we evaluate the model's capability for generalized instruction following by testing on out-of-domain datasets, novel prompts, and unseen tasks. Our empirical experiments reveal that our multi-task SpeechVerse model is even superior to conventional task-specific baselines on 9 out of the 11 tasks.

Rana Salama, Jason Cai, Michelle Yuan et al.

Large language model (LLM) agents have evolved to intelligently process information, make decisions, and interact with users or tools. A key capability is the integration of long-term memory capabilities, enabling these agents to draw upon historical interactions and knowledge. However, the growing memory size and need for semantic structuring pose significant challenges. In this work, we propose an autonomous memory augmentation approach, MemInsight, to enhance semantic data representation and retrieval mechanisms. By leveraging autonomous augmentation to historical interactions, LLM agents are shown to deliver more accurate and contextualized responses. We empirically validate the efficacy of our proposed approach in three task scenarios; conversational recommendation, question answering and event summarization. On the LLM-REDIAL dataset, MemInsight boosts persuasiveness of recommendations by up to 14%. Moreover, it outperforms a RAG baseline by 34% in recall for LoCoMo retrieval. Our empirical results show the potential of MemInsight to enhance the contextual performance of LLM agents across multiple tasks.

Raphael Shu, Nilaksh Das, Michelle Yuan et al.

AI agents powered by large language models (LLMs) have shown strong capabilities in problem solving. Through combining many intelligent agents, multi-agent collaboration has emerged as a promising approach to tackle complex, multi-faceted problems that exceed the capabilities of single AI agents. However, designing the collaboration protocols and evaluating the effectiveness of these systems remains a significant challenge, especially for enterprise applications. This report addresses these challenges by presenting a comprehensive evaluation of coordination and routing capabilities in a novel multi-agent collaboration framework. We evaluate two key operational modes: (1) a coordination mode enabling complex task completion through parallel communication and payload referencing, and (2) a routing mode for efficient message forwarding between agents. We benchmark on a set of handcrafted scenarios from three enterprise domains, which are publicly released with the report. For coordination capabilities, we demonstrate the effectiveness of inter-agent communication and payload referencing mechanisms, achieving end-to-end goal success rates of 90%. Our analysis yields several key findings: multi-agent collaboration enhances goal success rates by up to 70% compared to single-agent approaches in our benchmarks; payload referencing improves performance on code-intensive tasks by 23%; latency can be substantially reduced with a routing mechanism that selectively bypasses agent orchestration. These findings offer valuable guidance for enterprise deployments of multi-agent systems and advance the development of scalable, efficient multi-agent collaboration frameworks.

Yubin Ge, Salvatore Romeo, Jason Cai et al.

Temporal reasoning in multi-session dialogues presents a significant challenge which has been under-studied in previous temporal reasoning benchmarks. To bridge this gap, we propose a new evaluation task for temporal reasoning in multi-session dialogues and introduce an approach to construct a new benchmark by augmenting dialogues from LoCoMo and creating multi-choice QAs. Furthermore, we present TReMu, a new framework aimed at enhancing the temporal reasoning capabilities of LLM-agents in this context. Specifically, the framework employs time-aware memorization through timeline summarization, generating retrievable memory by summarizing events in each dialogue session with their inferred dates. Additionally, we integrate neuro-symbolic temporal reasoning, where LLMs generate Python code to perform temporal calculations and select answers. Experimental evaluations on popular LLMs demonstrate that our benchmark is challenging, and the proposed framework significantly improves temporal reasoning performance compared to baseline methods, raising from 29.83 on GPT-4o via standard prompting to 77.67 via our approach and highlighting its effectiveness in addressing temporal reasoning in multi-session dialogues.

Young Min Cho, Daniele Bonadiman, Divya Bhargavi et al.

Training large foundation models for agentic tasks is increasingly impractical due to the high computational costs, long iteration cycles, and rapid obsolescence as new models are continuously released. Instead of post-training massive models for every new task or domain, we propose Supplement Generation Training (SGT), a more efficient and sustainable strategy. SGT trains a smaller LLM to generate useful supplemental text that, when appended to the original input, helps the larger LLM solve the task more effectively. These lightweight models can dynamically adapt supplements to task requirements, improving performance without modifying the underlying large models. This approach decouples task-specific optimization from large foundation models and enables more flexible, cost-effective deployment of LLM-powered agents in real-world applications.

Suhaib Abdurahman, Etsuko Ishii, Katerina Margatina et al.

LLM-based multi-agent systems (MAS) show promise on complex tasks but remain prone to coordination failures such as goal drift, error cascades, and misaligned behaviors. We propose Explicit Trait Inference (ETI), a psychologically grounded method for improving coordination. ETI enables agents to infer and track partner characteristics along two established psychological dimensions--warmth (e.g., trust) and competence (e.g., skill)--from interaction histories to guide decisions. We evaluate ETI in controlled settings (economic games), where it reduces payoff loss by 45-77%, and in more realistic, complex multi-agent settings (MultiAgentBench), where it improves performance by 3-29% depending on the scenario and model, relative to a CoT baseline. Additional analysis shows that gains are closely linked to trait inference: ETI profiles predict agents' actions, and informative profiles drive improvements. These results highlight ETI as a lightweight and robust mechanism for improving coordination in diverse multi-agent settings, and provide the first systematic evidence that LLM agents can (i) reliably infer others' traits from interaction histories and (ii) leverage structured awareness of others' traits for coordination.

Sadat Shahriar, Zheng Qi, Nikolaos Pappas et al.

Aligning Large Language Models (LLM) to address subjectivity and nuanced preference levels requires adequate flexibility and control, which can be a resource-intensive and time-consuming procedure. Existing training-time alignment methods require full re-training when a change is needed and inference-time ones typically require access to the reward model at each inference step. To address these limitations, we introduce inference-time model alignment method that learns encoded representations of preference dimensions, called \textit{Alignment Vectors} (AV). These representations are computed by subtraction of the base model from the aligned model as in model editing enabling dynamically adjusting the model behavior during inference through simple linear operations. Even though the preference dimensions can span various granularity levels, here we focus on three gradual response levels across three specialized domains: medical, legal, and financial, exemplifying its practical potential. This new alignment paradigm introduces adjustable preference knobs during inference, allowing users to tailor their LLM outputs while reducing the inference cost by half compared to the prompt engineering approach. Additionally, we find that AVs are transferable across different fine-tuning stages of the same model, demonstrating their flexibility. AVs also facilitate multidomain, diverse preference alignment, making the process 12x faster than the retraining approach.

Tamer Alkhouli, Katerina Margatina, James Gung et al.

We introduce Conversational Function-Calling Evaluation Through Turn-Level Interactions (CONFETTI), a conversational benchmark1 designed to evaluate the function-calling capabilities and response quality of large language models (LLMs). Current benchmarks lack comprehensive assessment of LLMs in complex conversational scenarios. CONFETTI addresses this gap through 109 human-simulated conversations, comprising 313 user turns and covering 86 APIs. These conversations explicitly target various conversational complexities, such as follow-ups, goal correction and switching, ambiguous and implicit goals. We perform off-policy turn-level evaluation using this benchmark targeting function-calling. Our benchmark also incorporates dialog act annotations to assess agent responses. We evaluate a series of state-of-the-art LLMs and analyze their performance with respect to the number of available APIs, conversation lengths, and chained function calling. Our results reveal that while some models are able to handle long conversations, and leverage more than 20+ APIs successfully, other models struggle with longer context or when increasing the number of APIs. We also report that the performance on chained function-calls is severely limited across the models. Overall, the top performing models on CONFETTI are Nova Pro (40.01%), Claude Sonnet v3.5 (35.46%) and Llama 3.1 405B (33.19%) followed by command-r-plus (31.18%) and Mistral-Large-2407 (30.07%).

Ming Shen, Raphael Shu, Anurag Pratik et al.

We have seen remarkable progress in large language models (LLMs) empowered multi-agent systems solving complex tasks necessitating cooperation among experts with diverse skills. However, optimizing LLM-based multi-agent systems remains challenging. In this work, we perform an empirical case study on group optimization of role-based multi-agent systems utilizing natural language feedback for challenging software development tasks under various evaluation dimensions. We propose a two-step agent prompts optimization pipeline: identifying underperforming agents with their failure explanations utilizing textual feedback and then optimizing system prompts of identified agents utilizing failure explanations. We then study the impact of various optimization settings on system performance with two comparison groups: online against offline optimization and individual against group optimization. For group optimization, we study two prompting strategies: one-pass and multi-pass prompting optimizations. Overall, we demonstrate the effectiveness of our optimization method for role-based multi-agent systems tackling software development tasks evaluated on diverse evaluation dimensions, and we investigate the impact of diverse optimization settings on group behaviors of the multi-agent systems to provide practical insights for future development.

Karthikeyan K, Michelle Yuan, Elman Mansimov et al.

Recent works have demonstrated that incorporating search during inference can significantly improve reasoning capabilities of language agents. Some approaches may make use of the ground truth or rely on model's own generated feedback. The search algorithm uses this feedback to then produce values that will update its criterion for exploring and exploiting various reasoning paths. In this study, we investigate how search and model's self-feedback can be leveraged for reasoning tasks. First, we explore differences in ground-truth feedback and self-feedback during search for math reasoning. Second, we observe limitations in applying search techniques to more complex tasks like tool-calling and design domain-specific approaches to address these gaps. Our experiments reveal challenges related to generalization when solely relying on self-feedback during search. For search to work effectively, either access to the ground-truth is needed or feedback mechanisms need to be carefully designed for the specific task.

Michelle Yuan, Khushbu Pahwa, Shuaichen Chang et al.

Designing effective agentic systems requires the seamless composition and integration of agents, tools, and models within dynamic and uncertain environments. Most existing methods rely on static, semantic retrieval approaches for tool or agent discovery. However, effective reuse and composition of existing components remain challenging due to incomplete capability descriptions and the limitations of retrieval methods. Component selection suffers because the decisions are not based on capability, cost, and real-time utility. To address these challenges, we introduce a structured, automated framework for agentic system composition that is inspired by the knapsack problem. Our framework enables a composer agent to systematically identify, select, and assemble an optimal set of agentic components by jointly considering performance, budget constraints, and compatibility. By dynamically testing candidate components and modeling their utility in real-time, our approach streamlines the assembly of agentic systems and facilitates scalable reuse of resources. Empirical evaluation with Claude 3.5 Sonnet across five benchmarking datasets shows that our online-knapsack-based composer consistently lies on the Pareto frontier, achieving higher success rates at significantly lower component costs compared to our baselines. In the single-agent setup, the online knapsack composer shows a success rate improvement of up to 31.6% in comparison to the retrieval baselines. In multi-agent systems, the online knapsack composer increases success rate from 37% to 87% when agents are selected from an agent inventory of 100+ agents. The substantial performance gap confirms the robust adaptability of our method across diverse domains and budget constraints.

Yubin Ge, Salvatore Romeo, Jason Cai et al.

Despite the rapid advancements in LLM agents, they still face the challenge of generating meaningful reflections due to inadequate error analysis and a reliance on rare successful trajectories, especially in complex tasks. In this work, we propose SAMULE, a new framework for self-learning agents powered by a retrospective language model that is trained based on Multi-Level Reflection Synthesis. It first synthesizes high-quality reflections across three complementary levels: Single-Trajectory Learning (micro-level) for detailed error correction; Intra-Task Learning (meso-level) to build error taxonomies across multiple trials of the same task, and Inter-Task Learning (macro-level) to extract transferable insights based on same typed errors from diverse task failures. Then we fine-tune a language model serving as the retrospective model to generate reflections during inference. We further extend our framework to interactive settings through a foresight-based reflection mechanism, enabling agents to proactively reflect and adapt during user interactions by comparing predicted and actual responses. Extensive experiments on three challenging benchmarks - TravelPlanner, NATURAL PLAN, and Tau-bench - demonstrate that our approach significantly outperforms reflection-based baselines. Our results highlight the critical role of well-designed reflection synthesis and failure-centric learning in building self-improving LLM agents.

Young-Min Cho, Raphael Shu, Nilaksh Das et al.

Effective group decision-making is critical in Multi-Agent Systems (MAS). Yet, how different mechanisms for reaching consensus impact collaboration quality and efficiency remains understudied. We conduct a systematic study on group decision-making mechanisms in a decentralized setting. Through controlled experiments, we analyze how different voting rules affect decision quality and efficiency in a multi-round collaboration. Results reveal that majority voting often cause inefficient collaboration due to its strict acceptance criteria. At the extreme, unanimous voting gives 87% lower initial performance than the best-performing method. Our qualitative analysis of cross-agent communication shows that messages become longer and more repetitive over time: while message length increases by 84%, similarity to the previous round increases to 90%. Based on these insights, language-based early stopping methods make the performance 13% closer to oracle while reducing rounds by 50%. Our findings highlight the crucial role of group decision-making in optimizing MAS collaboration.

Jason Cai, Hang Su, Monica Sunkara et al.

Large Language Models (LLMs) are powerful models for generation tasks, but they may not generate good quality outputs in their first attempt. Apart from model fine-tuning, existing approaches to improve prediction accuracy and quality typically involve LLM self-improvement / self-reflection that incorporate feedback from models themselves. Despite their effectiveness, these methods are hindered by their high computational cost and lack of scalability. In this work, we propose CERET, a method for refining text generations by considering semantic stability, entailment and inter-sample uncertainty measures. Experimental results show that CERET outperforms Self-consistency and Self-rerank baselines consistently under various task setups, by ~1.6% in Rouge-1 for abstractive summarization and ~3.5% in hit rate for question answering. Compared to LLM Self-rerank method, our approach only requires 9.4% of its latency and is more cost-effective.

Jinglun Cai, Monica Sunkara, Xilai Li et al.

Masked Language Models (MLMs) have proven to be effective for second-pass rescoring in Automatic Speech Recognition (ASR) systems. In this work, we propose Masked Audio Text Encoder (MATE), a multi-modal masked language model rescorer which incorporates acoustic representations into the input space of MLM. We adopt contrastive learning for effectively aligning the modalities by learning shared representations. We show that using a multi-modal rescorer is beneficial for domain generalization of the ASR system when target domain data is unavailable. MATE reduces word error rate (WER) by 4%-16% on in-domain, and 3%-7% on out-of-domain datasets, over the text-only baseline. Additionally, with very limited amount of training data (0.8 hours), MATE achieves a WER reduction of 8%-23% over the first-pass baseline.

Nilaksh Das, Monica Sunkara, Sravan Bodapati et al.

End-to-end ASR models trained on large amount of data tend to be implicitly biased towards language semantics of the training data. Internal language model estimation (ILME) has been proposed to mitigate this bias for autoregressive models such as attention-based encoder-decoder and RNN-T. Typically, ILME is performed by modularizing the acoustic and language components of the model architecture, and eliminating the acoustic input to perform log-linear interpolation with the text-only posterior. However, for CTC-based ASR, it is not as straightforward to decouple the model into such acoustic and language components, as CTC log-posteriors are computed in a non-autoregressive manner. In this work, we propose a novel ILME technique for CTC-based ASR models. Our method iteratively masks the audio timesteps to estimate a pseudo log-likelihood of the internal LM by accumulating log-posteriors for only the masked timesteps. Extensive evaluation across multiple out-of-domain datasets reveals that the proposed approach improves WER by up to 9.8% and OOV F1-score by up to 24.6% relative to Shallow Fusion, when only text data from target domain is available. In the case of zero-shot domain adaptation, with no access to any target domain data, we demonstrate that removing the source domain bias with ILME can still outperform Shallow Fusion to improve WER by up to 9.3% relative.

Dhanush Bekal, Ashish Shenoy, Monica Sunkara et al.

Accurate recognition of slot values such as domain specific words or named entities by automatic speech recognition (ASR) systems forms the core of the Goal-oriented Dialogue Systems. Although it is a critical step with direct impact on downstream tasks such as language understanding, many domain agnostic ASR systems tend to perform poorly on domain specific or long tail words. They are often supplemented with slot error correcting systems but it is often hard for any neural model to directly output such rare entity words. To address this problem, we propose k-nearest neighbor (k-NN) search that outputs domain-specific entities from an explicit datastore. We improve error correction rate by conveniently augmenting a pretrained joint phoneme and text based transformer sequence to sequence model with k-NN search during inference. We evaluate our proposed approach on five different domains containing long tail slot entities such as full names, airports, street names, cities, states. Our best performing error correction model shows a relative improvement of 7.4% in word error rate (WER) on rare word entities over the baseline and also achieves a relative WER improvement of 9.8% on an out of vocabulary (OOV) test set.

Ashish Shenoy, Sravan Bodapati, Monica Sunkara et al.

Neural Language Models (NLM), when trained and evaluated with context spanning multiple utterances, have been shown to consistently outperform both conventional n-gram language models and NLMs that use limited context. In this paper, we investigate various techniques to incorporate turn based context history into both recurrent (LSTM) and Transformer-XL based NLMs. For recurrent based NLMs, we explore context carry over mechanism and feature based augmentation, where we incorporate other forms of contextual information such as bot response and system dialogue acts as classified by a Natural Language Understanding (NLU) model. To mitigate the sharp nearby, fuzzy far away problem with contextual NLM, we propose the use of attention layer over lexical metadata to improve feature based augmentation. Additionally, we adapt our contextual NLM towards user provided on-the-fly speech patterns by leveraging encodings from a large pre-trained masked language model and performing fusion with a Transformer-XL based NLM. We test our proposed models using N-best rescoring of ASR hypotheses of task-oriented dialogues and also evaluate on downstream NLU tasks such as intent classification and slot labeling. The best performing model shows a relative WER between 1.6% and 9.1% and a slot labeling F1 score improvement of 4% over non-contextual baselines.

Monica Sunkara, Chaitanya Shivade, Sravan Bodapati et al.

While there have been several contributions exploring state of the art techniques for text normalization, the problem of inverse text normalization (ITN) remains relatively unexplored. The best known approaches leverage finite state transducer (FST) based models which rely on manually curated rules and are hence not scalable. We propose an efficient and robust neural solution for ITN leveraging transformer based seq2seq models and FST-based text normalization techniques for data preparation. We show that this can be easily extended to other languages without the need for a linguistic expert to manually curate them. We then present a hybrid framework for integrating Neural ITN with an FST to overcome common recoverable errors in production environments. Our empirical evaluations show that the proposed solution minimizes incorrect perturbations (insertions, deletions and substitutions) to ASR output and maintains high quality even on out of domain data. A transformer based model infused with pretraining consistently achieves a lower WER across several datasets and is able to outperform baselines on English, Spanish, German and Italian datasets.

Monica Sunkara, Srikanth Ronanki, Dhanush Bekal et al.

In this work, we explore a multimodal semi-supervised learning approach for punctuation prediction by learning representations from large amounts of unlabelled audio and text data. Conventional approaches in speech processing typically use forced alignment to encoder per frame acoustic features to word level features and perform multimodal fusion of the resulting acoustic and lexical representations. As an alternative, we explore attention based multimodal fusion and compare its performance with forced alignment based fusion. Experiments conducted on the Fisher corpus show that our proposed approach achieves ~6-9% and ~3-4% absolute improvement (F1 score) over the baseline BLSTM model on reference transcripts and ASR outputs respectively. We further improve the model robustness to ASR errors by performing data augmentation with N-best lists which achieves up to an additional ~2-6% improvement on ASR outputs. We also demonstrate the effectiveness of semi-supervised learning approach by performing ablation study on various sizes of the corpus. When trained on 1 hour of speech and text data, the proposed model achieved ~9-18% absolute improvement over baseline model.

Monica Sunkara, Srikanth Ronanki, Kalpit Dixit et al.

Automatic speech recognition (ASR) systems in the medical domain that focus on transcribing clinical dictations and doctor-patient conversations often pose many challenges due to the complexity of the domain. ASR output typically undergoes automatic punctuation to enable users to speak naturally, without having to vocalise awkward and explicit punctuation commands, such as "period", "add comma" or "exclamation point", while truecasing enhances user readability and improves the performance of downstream NLP tasks. This paper proposes a conditional joint modeling framework for prediction of punctuation and truecasing using pretrained masked language models such as BERT, BioBERT and RoBERTa. We also present techniques for domain and task specific adaptation by fine-tuning masked language models with medical domain data. Finally, we improve the robustness of the model against common errors made in ASR by performing data augmentation. Experiments performed on dictation and conversational style corpora show that our proposed model achieves ~5% absolute improvement on ground truth text and ~10% improvement on ASR outputs over baseline models under F1 metric.